You are on page 1of 89

1

PUBLIC LAW (LPUB 4814)

STUDY MATERIAL

2018

Copyright vests in UFS and may not be infringed


2

NOTES:

ASPECTS OF THE
RIGHT TO EQUALITY

Compiler:

Prof. J.L. Pretorius


2007 (updated by Dr M.E. Marais 2018)

ASPECTS OF THE RIGHT TO EQUALITY


3

INDEX
UNIT 1

THE CONSTITUTIONAL COURT’S EQUALITY JURISPRUDENCE………………………..6


1. Introduction…………………………………………………………………... ………………….6
2. Substantive and formal equality………………………........................................................7
3. “Differentiation” and “discrimination”…………………………………………………………. 9
4. Requirements for mere differentiation …………………………………………… ………..10
4.1 Legitimate purpose………………………………………………….………………... 10
4.2 Rational connection…………………………………………………………………... 11
4.3 Justification of mere differentiation………………………………............................13
5. Unfair discrimination…………………………………………………………………….... 13
5.1 What is discrimination……………………………………………………………..… 13
5.2 When is discrimination “unfair”?.......................................................................... 15
5.2.1 Impact of the discrimination on the complainant……………………….. 15
5.2.2 Position of the complainants in society………………………………….. 16
5.2.3 Nature of the provision or power and the purpose
sought to be achieved by it……………………………………………….17
5.2.4 Extent to which the discrimination has affected
the rights or interests of the complainants ………………………………… 19
5.3 Justification of unfair discrimination……………………………………………… ……… 22
5.3.1 Proportionality……………………………………………………………………………. 22
5.3.2 Nature and importance of the right limited……....................................................... 23
5.3.3 Importance of the purpose of the limitation…………………………….................... 23
5.3.4 Nature and extent of the limitation……………....................................................... 23
5.3.5 Relation between the limitation and its purpose………………………................... 24
5.3.6 Less restrictive means to achieve the purpose…………………………………….. 25
6. Affirmative action……………………………………………………………………… 25

UNIT 2

GENERAL PRINCIPLES OF EMPLOYMENT DISCRIMINATION LAW


EMPLOYMENT EQUITY ACT………………………………………………………………… 27
A Introduction…………………………………………………………………………………… 27
B Prohibition of employment discrimination………………………………............................ 28
1. Provisions of the EEA………………………………………………………..........................28
2. Direct and indirect employment discrimination………………………...............................29
3. Proof of direct employment discrimination…………………………….............................. 30
4. Indirect employment discrimination…………………………………………………………. 33
4.1 Introduction………………………………………………………….......................... 33
4.2 Locus classicus……………………………………………………………………….. 33
4.3 Elements of indirect employment discrimination………………….........................34
4

(a) Neutral employment policy or practice as the cause of detriment to a category of people
protected under a prohibited ground (causality)………………………………..........................35
(b) Disproportional impact……………………………………….................................................36
(b.1) Identifying the pool of comparison…………………………………………………..36
(b.2) Establishing the relevant proportions……………………………………….………38
(b.3) Determining substantial disproportionality……..…………………………….. … 38
5. Justification of employment discrimination………………………………………………….….39
5.1
Introduction…………………………………………………………..............................................40
5.2 Inherent requirements of the
job……………………………………………………...................................................................40
(a) Meaning of inherent job requirements…………………………………………………………40
(b) Health and safety…………………………………………….................................................40
(c) Cost and profitability…………………………………………................................................43
(d) Legitimate third-party preferences……………………………………………………………. 44
5.3 Operational justifications………………………………………………………..……. 45
5.3.1 Introduction…………………………………………………………………. .45
5.3.2 Economic grounds of justification…………………………………….……46
5.3.3 Administrative burdens…………………………………............................48
5.3.4 Maintenance of sound labour relations /
avoidance of labour unrest………………………………………………………..50
5.3.5 Market forces…………………………………………………………………51
5.4 Public interest justification……………………………………………………..………51
6 Reasonable accommodation……………………………………………………………………..53
7 Harassment………………………………………………………..............................................53

UNIT 3

THE PROMOTION OF EQUALITY AND PREVENTION OF UNFAIR DISCRIMINATION ACT


1. Background……………………………………………………………….……………………….54
2. Standing………………………………………………………………………………………… 54
3. Application…………………………………………………………………………………………56
4. Jurisdiction………………………………………………………………………………………...57
5. Procedure…………………………………………………………………...……………………. 58
6. Remedies……………………………………………………………………………………… . 60
7. Appeal and review……………………………………………………………………………….. 62
8. Proof of unfair discrimination…………………………………………..……………………….. 62
8.1 Introduction…………………………………………………………………………… 62
8.2 Test for prima facie discrimination………………………………………………. 62
8.3 Burden of proof regarding the determination of discrimination….……………… 64
8.4 Proof of unfairness…………………………………………………………………… 65
8.4.1 Positive measures for protection and advancement…… …………………….. 66
8.4.2 The enquiry into unfairness in terms of section 14(2) and (3)……………………66
(a) Context…………………………………………………………………………...66
5

(b) Factors listed in section 14(3)………………………………………………....66


(c) The requirement of section 14(2)(c)…………………………………………..68

UNIT 4

HARASSMENT
1. Introduction…………………………………………………………………............................. 73
2. Sexual harassment…………………………………………………………………………… 74
2.1 What is sexual harassment?................................................................................ 74
2.2 Forms of sexual harassment……………………………………...............................75
(a) Quid pro quo harassment……………………………………....................... 76
(b) Hostile work environment……………………………………....................... 76
2.3 Elements of sexual harassment……………………………………………………... 77
(a) “Unwanted conduct”…………………………………………………………... 78
(b) “Of a sexual nature”………………………………....................................... 79
(c) Persistency or seriousness…………………………………………………... 80
(d) Impairment of dignity………………………………………………………….. 80
3. Racial, ethnic and national origin harassment…………………………… …………………. 81

UNIT 5

AFFIRMATIVE ACTION
Pretorius et al. Employment Equity Law Chapter 9 (See extract)

UNIT 6

REASONABLE ACCOMMODATION
Pretorius et al. Employment Equity Law Chapter 7(See extract)

UNIT 7

PREFERENTIAL PROCUREMENT AND BLACK ECONOMIC EMPOWERMENT


See Study Guide
6

UNIT 1

THE CONSTITUTIONAL COURT’S EQUALITY


JURISPRUDENCE

1. Introduction
Section 9 of the Constitution provides:

(1) Everyone is equal before the law and has the right to equal protection and benefit of the law.

(2) Equality includes the full and equal enjoyment of all rights and freedoms. To promote the
achievement of equality, legislative and other measures designed to protect or advance persons,
or categories of persons, disadvantaged by unfair discrimination may be taken.

(3) The state may not unfairly discriminate directly or indirectly against anyone on one or more
grounds, including race, gender, sex, pregnancy, marital status, ethnic or social origin, colour,
sexual orientation, age, disability, religion, conscience, belief, culture, language and birth.

(4) No person may unfairly discriminate directly or indirectly against anyone on one or more
grounds in terms of subsection (3). National legislation must be enacted to prevent or prohibit unfair
discrimination.

(5) Discrimination on one or more of the grounds listed in subsection (3) is unfair unless it is
established that the discrimination is fair.

The Constitutional Court has interpreted Section 9 in a number of cases. The most important of
these cases are:

 President of RSA v Hugo 1997 (6) BCLR 708 (CC)


 Prinsloo v Van der Linde 1997 (6) BCLR 759 (CC)
 Harksen v Lane 1997 (11) BCLR 1489 (CC).
 City Council of Pretoria v Walker 1998 (3) BCLR 257 (CC).
 National Coalition for Gay and Lesbian Equality v Minister of Home Affairs 2000 (1) BCLR
39 (CC).

For present purposes the two most important aspects of the Constitutional Court’s interpretation of
the equality clause are:

 The Court’s endorsement of a notion of substantive equality, and


 The distinction between “mere differentiation” and (unfair) “discrimination”.

These aspects will be considered next.


7

2. Substantive and formal equality


The Constitutional Court has, as is also the case in other jurisdictions like Canada and the
European Union, drawn a distinction between so-called formal and substantive equality. Stated
simply, formal equality means that people and groups of people should be treated identically,
regardless of their particular circumstances. Formal equality simply requires that people should
have the same rights and entitlements. Equality is achieved if people are treated in a neutral
fashion, i.e. that they should be afforded the same rights and entitlements, without taking their
particular social and historical circumstances into account. Equality is therefore achieved if people
are afforded the opportunity to compete equally on the basis of individual talents. It is assumed
that as individuals, everyone is basically similarly placed and that an application of the law in
accordance with a standard of strict neutrality best serves to avoid unjust preference or prejudice.

The formal equality approach does not take the social and economic disparities between groups
and individuals into consideration. Those who ascribe to a substantive equality approach, on the
other hand, insist that equality can only be achieved if people’s actual social and economic
conditions are examined. They therefore dispute the basic starting point of the formal equality
approach, namely that equality only requires people to be treated in a similar fashion.

It is argued that the formal equality approach fails to recognise that there are underlying patterns
of group-based disadvantage which are not adequately addressed if people are merely treated
similarly. Similarity of treatment may, in certain circumstances, reinforce rather than address social
disadvantage. Only by examining the larger social context can a court determine if differential
treatment will cause inequality, or whether it would be identical treatment, which would in the
particular context result in equality or foster disadvantage.

The Constitutional Court has held that the wording of the equality clause of the Constitution makes
a clear choice in favour of the substantive approach.1 In National Coalition for Gay and Lesbian
Equality v Min. of Justice 1998 (12) 1517 (CC) Ackerman J observed that it is insufficient for the
Constitution to merely ensure that discriminatory treatment of the past is eliminated. He
emphasised that past discriminatory actions may therefore require special measures in order to
level the playing field between individuals and groups. By merely treating them identically, this may
not be achieved.

The substantive equality approach is not always easy to apply and would require from courts of
law particular sensitivity to the larger socio-historical position of groups. The following two
examples will give an idea how the Court has tried to apply the substantive equality approach, to
particular South African situations.

In President of the RSA v Hugo (supra), the Court had to consider the constitutionality of the
decision of the President in terms of which certain categories of women prisoners with children

1 See National Coalition for Gay and Lesbian Equality v Min. of Justice 1998 (12) 1517 par 62.
8

under the age of 12 were released from prison. No men in similar circumstances were released.
The Court held that the remission of sentence did not unfairly impact on the rights and interests of
male prisoners in comparable circumstances. It was noted that since the vast majority of prisoners
were men, a remission of sentence in their case would have very different consequences. The
extent to which the rights of male prisoners had been affected was also not substantial. They were
merely deprived of an early release to which they had no legal entitlement. Moreover, they were
not prevented from applying directly to the President for remission of sentence on an individual
basis in the light of their own special circumstances. The Court took into account that, generally
speaking, men and women are not similarly situated in respect of their child-caring responsibilities.
Social conditions are still such that this obligation is primarily borne by women.

In City Council of Pretoria v Walker (supra) the Court had to decide whether different municipal
rates for water and electricity which applied in the historically white and black parts of the city were
unconstitutional. In the historically black parts residents had to pay a so-called “flat rate”, i.e. a
fixed amount irrespective of the actual consumption of services. In the historically white areas, the
rates were consumption-based. The flat rate was calculated on the basis of the average
consumption of water and electricity per household in the respective black areas. As the Court
observed, this was a crude method of recovering charges; it meant that those residences who
consumed less water and electricity than the average resident of Atteridgeville and Mamelodi paid
the same as those whose consumption was above the average. It also meant that if consumption
in the two townships increased, the flat rate would be inadequate to recover the tariff charges for
the actual consumption. This system was a relic of the past system of racial segregation of
residential areas and the lack of development of black residential areas. Previously, no meters
had been installed in Atteridgeville and Mamelodi to measure consumption of water and electricity.
In practice, there was no other way to recover water and electricity charges except by means of
the flat rate. After the coming into effect of the interim constitution, the City Council started a
process of installing meters in the historically black townships. This process had, however, not
been completed at the time of the judgement. A group of white residents instituted proceedings,
claiming that the differential system violated their right to equality. They argued that because the
flat rate was lower than a consumption-based rate, the residents of the predominantly white parts
of Pretoria were subsiding those of Mamelodi and Atteridgeville. From a formal equality viewpoint,
the lack of a uniform rating system would clearly be discriminatory. However, once again the Court
chose to follow a substantive approach rather than a formal one. It argued that there might well
be cases where it is not unfair to charge according to different rates for the same services. The
Court took into account the following considerations. The differential rating system was inherited
by the Council and could not be immediately eliminated. The operation of the flat rate in the
historically black areas was a temporary measure, which would last only as long as the process of
installing meters to measure water and electricity consumption in the black areas had been
completed. Moreover, there was no evidence that the respondents had been adversely affected
in any material way by the flat rate. There was no evidence of any deterioration in the high standard
of delivery of services in the historically white part of Pretoria since the amalgamation of the
previously separate residential areas. On the other hand, although there was some improvement
in the two black townships, the evidence showed that service delivery was still not satisfactory. It
was also not practical or feasible to have applied the flat rate throughout the municipality, since
9

this would have resulted in even greater disparity between individual ratepayers, than the present
system. The Court concluded that under the circumstances the adoption of a flat rate as an interim
arrangement while meters were being installed in the residential areas of the townships was the
only practical solution on the problem.

Equality and human dignity

An important feature of the Constitutional Court’s interpretation of equality is that it has stressed
the link between equality and human dignity. Discrimination is seen as a form of violation of a
person’s dignity. (See National Coalition for Gay and Lesbian Equality v Min of Justice: 1998 (12)
1517 par 126.)

3. “Differentiation” and “discrimination”


The Constitutional Court distinguishes between two forms of differential treatment, namely
“differentiation” and (unfair) “discrimination”:

Discrimination

“Discrimination” concerns all forms of differential treatment which-

(i) are based on any of the grounds specified in section 9(3), or

(ii) impair the dignity of a person, or

(iii) in some other invidious way adversely affect the complainant in a comparably serious
manner.

Discrimination can be direct or indirect. Indirect discrimination occurs when, though the
basis of a differentiation is neutral on its face, it has a discriminatory effect.

Mere differentiation

All other forms of differential treatment are considered to be “mere differentiation”.

The practical importance of the distinction is that the Court considers instances of discrimination
as a more serious form of unequal treatment than mere differentiation. In order to be
constitutionally valid, mere differentiation has only to be rational, i.e. there has to be a rational
relationship or connection between the differentiation and its purpose.2 If the differential treatment,
however, amounts to a “discrimination”, it has to meet, apart from the rationality requirement,

2 1997 (6) BCLR 759 (CC) at par 26


10

also the stricter demands of fairness under s 9(3). The requirements for both forms of differential
treatment will be discussed next.

4. Requirements for mere differentiation


In Harksen v Lane (supra) it was said:

“where section 8 [of the interim Constitution and now section 9 of the 1996 Constitution] is
invoked to challenge a legislative provision or executive conduct on the basis that it
differentiates between people or categories of people in a manner which amount to unequal
treatment or unfair discrimination, the first enquiry must be directed to the question whether
the impugned provision does differentiate between people or categories of people. If it does
so differentiate then in order not to fall foul of section 8(1) of the interim Constitution [or 9(1)
of the 1996 Constitution] there must be a rational connection between the differentiation
in question and the legitimate government purpose it is designed to achieve. If it is
justified in that way, it does not amount to a breach of section 8(1) [or 9(1)].”3

This means that in order to be “rational”, differential treatment must meet two requirements,
namely-

 it must serve a legitimate purpose, and

 there must be a rational connection between the differentiation and its purpose.

4.1. Legitimate purpose

This does not seem to be a difficult requirement to meet. An indication of what is required for a
purpose to be legitimate appears from Prinsloo v Van der Linde (supra). It was held that the
purpose of the requirement of rationality is to “promote the need for governmental action to relate
to a defensible vision of the public good”.4 According to Yacoob J in Jooste v Score Supermarket
Trading (Pty) Ltd, rationality review only involves an inquiry into “whether the differentiation is
arbitrary or irrational, or manifests naked preference.”5 In Municipality of Port Elizabeth v Rudman
the Court explained that the rational connection test only required that the “purpose of the
differentiation must not be arbitrary or capricious”.6

The notion of rationality espoused by the Constitutional Court resembles the minimal scrutiny
standard applied by American courts for the review of laws and regulations not involving a
fundamental constitutional right, suspect classification or the characteristics of citizenship, gender
or illegitimacy. In applying the rationality standard to the question of the legitimacy of governmental
purposes, the American courts have proved to be highly accommodating regarding the question
which governmental purposes would provide a legitimate basis for differentiation. In the view of

3 1997 (11) BCLR 1489 (CC) at 1506F-G. Brackets added.


4 1997 (6) BCLR 759 (CC) at par 25.
5 1999 (2) BCLR 139 (CC) at par 16.

6 1998 (4) BCLR 451 (SE) at 461G.


11

one commentator, “courts have traditionally exhibited extreme deference to the legislative definition
of the ‘general good’, either out of judicial sympathy for the difficulties of the legislative process, or
out of a belief in judicial restraint generally.”7

Examples

In Harksen v Lane8, the court considered the constitutionality of s 21 of the Insolvency Act 24 of
1936. In terms of this provision, the property of the spouse of a person whose estate has been
sequestrated automatically vests in the Master of the High Court and afterwards in the trustee. In
order to have the property released, the affected spouse must prove his or her ownership. The
majority judgement conceded the harmful effects of this procedure on the spouses of insolvents,
but nevertheless concluded that the measure was saved by a legitimate governmental interest
independent of its possible discriminatory consequences, namely to safeguard the interests of
creditors of the insolvent estate.

The exclusion of foreigners from permanent teaching posts is related to the legitimate
governmental purpose to reduce unemployment among South African citizens, “particularly when
thousands of qualified educators are unemployed”: Larbi-Odam and Others v Member of the
Executive Council for Education (N-W Prov) and Another 1997 (12) BCLR 1655 (CC) at 1669E-F.

The monopoly of the Post Office on the delivery of postal services in terms of section 7 of the Post
Office Act 44 of 1958, serves the legitimate purpose of providing postal service for the public
as a whole at uniform and affordable rates: Van Rensburg v South African Post Office Ltd 1998
(10) BCLR 1307 (E) at 1318E-G.

4.2. Rational connection

This requirement entails that the differentiation must be logically connected to its purpose. In
Harksen v Lane (supra) it was stated that a rational connection exists if the differentiation is an
“appropriate and effective” means to achieve the measure’s legitimate objective.9 Thus if a
differentiation does not actually serve to promote its purported purpose, it will be held to be
irrational or arbitrary. In Harksen, the Court argued that the fact that section 21 of the Insolvency
Act may cause inconvenience, potential prejudice, embarrassment and even have drastic
consequences to the spouse of an insolvent, does not mean that it is arbitrary or without rationality.
It was of the opinion that the legislature acted rationally in taking the view that the common law
and the statutory remedies relating to impeachable transactions were insufficient to enable the
Master or the trustee to ensure that all the property of the insolvent spouse found its way into the
insolvent estate.10

Examples

7 Tribe American Constitutional Law 1440.


8 1997 (11) BCLR 1489 (CC).
9 1997 (11) BCLR 1489 (CC) at 1513C.

10 1997 (11) BCLR 1489 (CC) at 1513F-1514A.


12

In Prinsloo v Van der Linde (supra) the court considered the provisions of s 84 of the Forest Act
122 of 1984. This section provided that if damage is caused by a veld or bush fire which started on
land situated outside a fire control area, it is presumed that the damage was caused by the
negligence of the owner of the land. No such presumption obtained in respect of land situated
inside a fire control area. The Act thus differentiated between owners of land in fire control areas
and non-fire control areas. According to the Court, the differentiation between landowners in
controlled and non-controlled areas was rationally connected to the legitimate governmental
purpose of preventing veld fires. Owners of land in non-controlled areas were not required to
embark on fire control measures, but were encouraged to do so by a number of measures, one of
which was s 84, which created the presumption of negligence against such landowners if the fire
originated on their land. S. 84 therefore serves to increase the vigilance of those responsible for
land outside fire control areas.

Fraser v Children’s Court, Pretoria North and Others11 concerned the statutory provision which
required the consent of the biological mother of a child born out of wedlock, but not that of the
father, for its adoption. It was argued that this measure served the purpose of the best interest of
the child. The court argued, however, that in the context of an adoption statute where the real
concern of the law is whether an order for the adoption of the child is justified, a right to veto the
adoption based on the marital status of the parent was clearly an inappropriate concern that could
lead to unfair anomalies. For instance, the court pointed out that consent of the father would be
unnecessary even in a situation where the child has the strongest bonds with the father and the
mother has not shown the slightest interest in the nurturing and development of the child.

It must be well understood that the fact that the differentiation is not the least onerous or
burdensome means of achieving the purpose does not make it irrational. As is the case with
the legitimate purpose requirement, this is not a difficult requirement to meet. In Prinsloo v Van der
Linde the Court distinguished between the rationality test and the proportionality test under the
limitation clause (s 36 of the Constitution) in this regard. It was contended on behalf of the applicant
that section 84 of the Forest Act lacked rationality because the apparent object the legislature
sought to achieve by reversing the general rule regarding the incidence of onus, namely that
whoever avers must prove, could have been achieved by means of common law aids to proof,
such as the concept of res ipsa loquitur and the practice of triers of fact to require less evidence to
establish a prima facie case if the facts in issue are particularly within the knowledge of the
opposing party.12 The court held that the question whether the legislation could have been tailored
in a different and more acceptable way is relevant for the issue of justification, but irrelevant to the
question of whether there is a sufficient relationship between the means chosen and the end
sought. Thus, as long as the imposition of the onus is not arbitrary, there is no breach of s 8(1) of
the interim Constitution (now s 9(1) of the 1996 Constitution). Burdening landowners in non-
controlled areas with the onus of proof serves the legitimate purpose of prevention of veldt fires,
since it entails an inducement “for those responsible for land in non-controlled areas to be

11 1997 (2) BCLR 153 (CC) at 164F-G.


12 1997 (6) BCLR 759 (CC) at 768G-H.
13

specifically vigilant lest they find themselves saddled with the responsibility for damages caused
by fire spreading from their land.”13

It also follows that under- or over-inclusiveness as such does not render a measure arbitrary.
In President of the Republic of South Africa v Hugo the reasons given by the President for limiting
the special remission of sentence to imprisoned mothers with small children, was that it would
serve the interests of children, since as a general rule, mothers are primarily responsible for the
care of young children in our society.14 Whilst “many fathers play only a secondary role in child
rearing, the release of male prisoners would not have contributed as significantly to the
achievement of the President’s purpose as the release of mothers.” The Court thus accepted the
rational connection between the reasons for and the purpose of limiting the remission of sentence
to female prisoners with small children, in spite of the fact that not all imprisoned mothers in actual
fact were the primary caretakers of their young children, while at least some imprisoned fathers
were.

As will appear later, the existence of a less burdensome alternative as well as under- or over-
inclusiveness, on the other hand, is directly significant when testing the fairness of a
discrimination. As we said earlier, the fact that discrimination is considered to be a more serious
form of differential treatment, means that it is subjected to more stringent requirements in order to
be constitutionally valid. It is therefore understandable that if there is less burdensome (or
discriminatory) means available to achieve the purpose of the discrimination, it will be held to be
unfair.

4.3 Justification of mere differentiation

Although it is difficult to see how an arbitrary differentiation can nevertheless be justifiable under
the limitation clause, it is at least theoretically possible that countervailing considerations may be
so forceful as to justify even arbitrariness. The same factors as those discussed in respect of the
justification of unfair discrimination must be applied when considering the justifiability of section
9(1) infringements and will be discussed infra.

Read: Van der Walt v Metcash Trading 2002 (5) BCLR 454 (CC).

5. Unfair discrimination
5.1. What is discrimination?

As was indicated above, “discrimination” concerns forms of differentiation which-

 Are based on a ground specified in section 9(3), or


 Are not based on a specified ground, but either

13 Prinsloo v Van der Linde 1997 (6) BCLR 759 (CC) at 776B-C.
14 1997 (6) BCLR 708 (CC) at 727F-H.
14

- violate the dignity of a person, or


- affect a person in a comparably serious way.

Specified grounds:

The prohibited grounds of discrimination mentioned in the section 9(3) are race, sex, gender,
pregnancy, marital status, ethnic or social origin, colour, sexual orientation, age, disability, religion,
conscience, belief, culture, language and birth.

In the case of discrimination based on a specified ground, section 9(5) applies, which provides that
discrimination on one or more of the grounds listed in subsection (3) is unfair unless it is established
that the discrimination is fair. No such presumption in favour of unfairness applies to discrimination
based on unspecified grounds.

Non-specified grounds:

The Constitutional Court has not endeavoured to provide a definition of “discrimination which has
the potential to impair the fundamental dignity of persons”, or given an indication of which forms of
discrimination will qualify as “discrimination with the potential to affect persons in a comparably
serious manner”. So far the Court has limited these categories to discrimination based on
unspecified grounds of discrimination that are analogous to the specified ones. The Court has
held that the listed grounds relate to attributes or characteristics that impact on human dignity.15
Analogous grounds will therefore have a similar relationship and impact.

The Court, however, has expressly cautioned against too narrow a definition of the terms “attributes
or characteristics”:

“[s]ection 8(2) of the interim Constitution seeks to prevent the unequal treatment of people
based on such criteria which may, amongst other things, result in the construction of patterns
of disadvantage such as has occurred only too visibly in our history.”16

Examples

The distinction between owners or occupiers of land in fire control areas and those who own or
occupy land outside of those areas in terms of the Forest Act, cannot “by any stretch of the
imagination, be seen as impairing the dignity of the owner or occupier of land outside such areas.
There is likewise no basis for concluding that the differentiation in some other invidious way
adversely affects such owner or occupier in a comparably serious manner.”17

A differentiation based on the attribute of citizenship, however, has the potential to impair the
fundamental human dignity of persons as human beings. Foreign citizens are a minority in any
country, without “political muscle”. Citizenship is also a personal attribute which is difficult to
change. The vulnerable position of foreign citizens in the specific circumstances of the case was

15 Harksen v Lane 1997 (11) BCLR 1489 (CC) at 1509E-G.


16 Id 1509E-G.
17 Prinsloo v Van der Linde 1997 (6) BCLR 759 (CC) at 776G-H.
15

further borne out by their exposure to the threats and intimidation concerning their appointment as
teachers.18

Differentiation based on marital status or relationships deemed to be marriage in terms of


section 21 of the Insolvency Act could have the potential to demean persons in their inherent
humanity and dignity.19 In Harksen v Lane (supra), O’Regan J relied in her dissenting judgement
on Miron v Trudel20 for a finding that marital status is closely related to human dignity and liberty.
In that case the Canadian Supreme Court held that “discrimination on the basis of marital status
touches the essential dignity and worth of the individual in the same way as other recognised
grounds of discrimination violative of fundamental human rights norms. Specifically, it touches the
individual’s freedom to live life with the mate of one’s choice. This is a matter of defining importance
to individuals. It is not a matter which should be excluded from Charter consideration on the ground
that its recognition would trivialise the equality guarantee.”

In Pretoria City Council v Walker the Constitutional Court stated that the fact that the differential
treatment was made applicable to geographical areas did not change the fact that in the
circumstances of the case it constituted discrimination, albeit indirect, on the grounds of race. It
would be artificial to make a comparison between an area known to be overwhelmingly a “black
area” and another known to be overwhelmingly a “white area” on the grounds of geography alone.
The effect of apartheid laws was that race and geography were inextricably linked. In the present
matter, the application of the geography standard, although seemingly neutral, in substance
differentiated between black and white residents.21

5.2. When is discrimination “unfair”?

5.2.1 Impact of the discrimination on the complainant

In accordance with the Constitutional Court’s adoption of the notion of substantive equality, the
fairness enquiry focuses primarily on the experience of the victim of discrimination. In the final
analysis it is the impact of the discrimination on the complainant that is the determining factor
regarding the unfairness of the discrimination.22

How does a court decide that a discriminatory provision has an unfair impact?

In order to determine whether the discriminatory provision has unfairly impacted on the
complainants, various factors must be considered, namely-

 The position of the complainants in society,

18 Larbi-Odam and Others v Member of the Executive Council for Education (N-W Prov) and Another 1997
(12) BCR 1655 (CC) at 1665F-1666C.
19 Harksen v Lane 1997(11) BCLR 1489 (CC) at 1515B. Marital status was not a listed ground in terms of

section 8(2) of the interim Constitution, but is now in terms of section 9(3) of the 1996 Constitution.
20 (1995) 29 CRR (2d) 189.

21 City Council of Pretoria v Walker 1998 (3) BCLR 257 (CC) at par 32.

22 Harksen v Lane 1997 (11) BCLR 1489 (CC) at 1510E.


16

 The nature of the provision or power and its purpose, and

 The extent to which the discrimination has affected the rights or interests of the complainant.

These factors do not constitute a closed list and it is the cumulative effect of these factors that must
be examined in order to make a determination regarding unfairness.23

5.2.2 Position of the complainant in society

What is of importance here is the “interplay between the discriminatory measure and the person or
group affected by it.”24 The more vulnerable the group adversely affected by the discrimination,
the more likely the discrimination will be held to be unfair.25 Relevant considerations include
whether they have suffered in the past from patterns of disadvantage, or whether the discrimination
in question is on a specified ground or not.26

In Brink v Kitshoff NO O’Regan J remarked that:

“[s]ection 8 [of the interim Constitution] was adopted then in the recognition that discrimination
against people who are members of disadvantaged groups can lead to patterns of group
disadvantage and harm. Such discrimination is unfair: it builds and entrenches inequality amongst
different groups in our society. The drafters realised that it was necessary both to proscribe such
forms of discrimination and to permit positive steps to redress the effects of such discrimination. The
need to prohibit such patterns of discrimination and to remedy their effects are the primary purposes
of section 8 and, in particular, subsection (2), (3) and (4).” 27

Which groups are considered vulnerable?

In the Hugo case, for example, the purpose of the Presidential Act to reduce the sentences of
certain groups of prisoners was to benefit three groups of prisoners, namely disabled prisoners,
young people and mothers of young children, as an act of mercy. The fact that all these groups
were regarded as being particularly vulnerable in our society, and that in the case of the disabled
and young mothers, they belonged to groups who had been victims of discrimination in the past,
weighed with the Court in concluding that the discrimination was not unfair.28

In the Hugo case, it was also held that imprisoned fathers of young children do not belong to a
class which had historically been disadvantaged.29

23 Harksen v Lane 1997 (11) BCLR 1489 (CC) at 1511C.


24 City Council of Pretoria v Walker 1998 (3) BCLR 257 (CC) at 279A-B.
25 President of the Republic of South Africa and Another v Hugo 1997 (6) BCLR 708 (CC) at 755E-F.

26 Harksen v Lane 1997 (11) BCLR 1489 (CC) at 1510F.

27 1996 (6) BCLR 752 (CC) at 769B-D.

28 1997 (6) BCLR 708 (CC) at par 47.

29 President of the Republic of South Africa and Another v Hugo 1997 (6) BCLR 708 (CC) at 728F-G.
17

Foreign citizens on the other hand are a vulnerable group, exposed to threats and intimidation
and lacking the kind of political influence that would induce governments to be more attentive of
their interests.30

The same applies to white municipal ratepayers, according to City Council of Pretoria v Walker.
Although in an economic sense this group is “neither disadvantaged nor vulnerable”, they are part
of a “racial minority which could, in a political sense, be regarded as vulnerable.”31

Requiring a judge’s certificate as a condition for unrepresented prisoners to prosecute an appeal


against their conviction or sentence imposed by a lower court, but not in the case of other
prospective appellants, imposes an extra handicap on a class of people “who harbour under the
greatest disadvantage in managing their appeals.”32

Gays and lesbians were also described in National Coalition for Gays and Lesbian Equality v
Minister of Justice and Others as a “political minority not able on their own to use political power to
secure favourable legislation for themselves”33 and a “permanent minority in society” who have
suffered in the past from patterns of disadvantage.

In any event, the Court has stressed that members of no racial group should be made to feel that
they are not deserving of equal “concern, respect and consideration” and that the law is likely to
be used against them more harshly that others who belong to other race groups. Although the
place of a complainant in the structures of advantage and disadvantage will always be one of the
central elements in the determination of how fair or unfair the challenged discrimination is,

“the doors of the courts must, of course, be equally open to all South Africans, independently of
whether historically they have been privileged or oppressed. Indeed, minorities of any kind are
always potentially vulnerable. Processes of differential treatment which have the legitimate purpose
of bringing about real equality should not be undertaken in a manner which gratuitously and
insensitively offends and marginalises persons identified as belonging to groups who previously
enjoyed advantage. Thus persons who have benefited from systemic advantage in the past and
who continue to enjoy such benefits today, are by no means excluded from the protection offered by
section 8.”

5.2.3 Nature of the provision or power and the purpose sought to be achieved by it

In order to be fair, discrimination must serve an important and worthy goal. In the Harksen case
the Court explained that —

“if the purpose is manifestly not directed, in the first instance, at impairing the rights of the
complainants, but is aimed at achieving a worthy and important societal goal, such as, for example,

30 Larbi-Odam and Others v Member of the Executive Council for Education (N-W Prov) and Another 1997
(12) BCLR 1655 (CC) at 1665G-H, 1666D-G.
31 1998 (3) BCLR 257 (CC) at 279G-280C. See also East London Transitional Local Council v Tax Payers

Action Organisation and Other 1998 (10) BCLR 1221 (E) at 1233D.
32 S v Ntuli 1996 (1) BCLR 141 (CC) at 151B.

33 1998 (12) BCLR 1517 (CC) at 1535E.


18

the furthering of equality for all, this purpose may, depending on the facts of the particular case, have
a significant bearing on the question whether complainants have in fact suffered the impairment in
question.”34

Since unfair discrimination deals with forms of discrimination affecting the dignity of
individuals or which is harmful in a comparably serious manner, only purposes of a more

substantial nature can establish fairness.35

Examples of purposes held to be important enough to make discrimination fair:

In National Coalition for Gay and Lesbian Equality and Others v Minister of Justice36 Heher J
concluded as follows in respect of the question whether “historical antipathy, personal
revulsion, religious conviction, the prevailing opinion in society, and the protection of the
morals of people” are considerations important enough to justify the common law crime of
sodomy: “[t]he first three must give way before the tenets of our Constitution which creates new
fundamental rights and recognise and protect old ones, and the last, while no doubt being a factor
properly to be considered in a given case, does not carry great weight where the law adequately
protects the vulnerable as it does in the case of possible homosexual (and heterosexual)
exploitation, by, for example, prosecutions for indecent assault or public indecency or
contravention of section 14 or 15 of the Sexual Offences Act.”

The control of possession and use of dependence-producing substances in terms of section


4(b) of the Drugs and Drug Trafficking Act 140 of 1992, qualify as such a “worthy object”.37

The prevention of collusion between spouses to the disadvantage of the creditors of the
insolvent spouse is an objective consistent with the values protected by the equality clause.38

In Van Rensburg v South African Post Office Ltd it was held that the purpose of the Post Office Act
is to provide for a postal service for the benefit of the public as a whole. The monopoly granted
to the Post Office in the provision of postal services was within the ambit of this purpose, since it
allows the Post Office to charge uniform affordable rates for the dispatch and delivery of post
throughout the country.39

5.2.4 Extent to which the discrimination has affected the rights or interests of the complainants

The more invasive the nature of the discrimination upon the interests of the affected party
is, the more likely it will be held to be unfair.40 Similarly, a factor tilting the scales towards a

34 1997 (11) BCLR 1489 (CC) at 1510G-H.


35 National Coalition for Gay and Lesbian Equality v Minister of Home Affairs 1999 (3) BCLR 280 (C) at 292I,
293C.
36 1998 (6) BCLR 726 (W) at 746I-747B.

37 Prince v President of the Law Society, Cape of Good Hope and Others 1998 (8) BCLR 976 (C) at 990I

38 Harksen v Lane 1997 (11) BCLR 1489 (CC) at 1515G, 1521F.

39 1998 (10) BCLR 307 (E) at 1318F-G.

40 President of the Republic of South Africa and Another v Hugo 1997 (6) BCLR 708 (CC) at 755F.
19

finding of unfairness will be whether the discrimination has led to an impairment of the
complainants’ fundamental dignity or an impairment of a comparably serious nature.41 For
instance, the Presidential Act in terms of which only female prisoners with young children were
granted a remission of sentence, did not unfairly impact on the rights and interests of male
prisoners in comparable circumstances. The Act merely deprived them of an early release to which
they had no legal entitlement. Moreover, they were not precluded from applying directly to the
President for remission of sentence on an individual basis in the light of their own special
circumstances.42

Some further examples:

In holding that the impact of the vesting of the property of the spouses of insolvents in the Master
or trustee in terms of section 12 of the Insolvency Act does not unfairly impact on the interests of
solvent spouses, the court in the Harksen case referred to a number of factors, such as that the
statutory vesting of the property of the solvent spouse does not have as consequence that such
property is necessarily removed from the possession of the solvent spouse. Only where the solvent
spouse claims property as his or hers and fails to adduce evidence to establish that claim on a
balance of probabilities then the insolvent estate is entitled to the property.43

The effect of excluding foreigners from permanent teaching positions unfairly affects the interests
of especially foreigners to whom the right of permanent residence has been granted. It impinges
on a vital interest, namely employment opportunities. Permanent residents have a right to compete
with South Africans in the employment market and it makes little sense to withhold employment
opportunities from those who have been allowed to stay permanently in the country.44

The discriminatory prohibition of sexual relations between gay men reinforces already existing
societal prejudices and severely increases the negative effects of such prejudices on their lives. It
perpetuates their social stigmatisation and encourages discrimination in employment, insurance
and judicial decisions about custody. Apart from the harm to their dignity and self-esteem, it
impinges on gay men in other harmful ways too, such as legitimising or encouragement of
blackmail, police entrapment, violence, refusal of facilities, accommodation and opportunities.45

General illustration of the principles of unfair discrimination developed by the court

41 Harksen v Lane 1997 (11) BCLR 1489 (CC) at 1511B, 1516G.


42 President of the Republic of South Africa and Another v Hugo 1997 (6) BCLR 708 (CC) at 732B-E, 756C-
D.
43 Harksen v Lane 1997 (11) BCLR 1489 (CC) at 1516A-C.

44 Larbi-Odam and Others v Member of the Executive Council for Education (N-W Prov) and Another 1997

(12) BCLR 1655 (CC) at 1667E-1668D.


45 National Coalition for Gay and Lesbian Equality v Minister of Justice and Others 1998 (12) BCLR 1517

(CC) at 1534D-1535D.
20

The following discussion is an illustration of the application of the principles in respect of fairness
of discriminatory action. It is an extract taken from De Waal, Currie and Erasmus Bill of Rights
Handbook (2000).46

To illustrate the difference between unfair discrimination and discrimination that is not unfair, let us
again look at the case of Pretoria City Council v Walker. As we said earlier, in this case, the
Constitutional Court examined the impact of two policies of the Pretoria City Council. The Council
had jurisdiction over the formerly exclusively white areas of Pretoria (‘old Pretoria’) and over the
townships of Atteridgeville and Mamelodi. The residents of old Pretoria were mostly white and
those of the two townships were mostly black. In old Pretoria ratepayers paid consumption-based
tariffs for the water and electricity services supplied by the Council. Actual consumption was
measured by meters placed in each property. In Atteridgeville and Mamelodi users paid a flat rate
per household, not matter how much or how little water or electricity they consumed. Walker, a
resident of old Pretoria, complained that the flat rate in Mamelodi and Atteridgeville was lower that
the metered rate and this therefore meant that the residents of old Pretoria subsidised those of the
two townships. He also complained that only residents of old Pretoria were singled out by the
council for legal action to recover arrears owed for services whilst a policy of non-enforcement was
followed in respect of Mamelodi and Atteridgeville

The majority of the Constitutional Court considered the actions of the Council to be indirect
discrimination on the listed ground of race. However, the majority went on to hold that the first set
of actions that Walker complained of (the flat rate and cross-subsidisation) was not unfair
discrimination while the second set (selective recovery of debts) was unfair discrimination. Unfair
discrimination is differentiation that has an unfair impact on its victims. In this regard, the Court first
took into account that Walker was white, and therefore belonged to a group that had not been
disadvantaged by the racial policies and practices of the past. In an economic sense, his group
was neither disadvantaged nor vulnerable, having been benefited rather than adversely affected
by discrimination in the past. What of the purpose of the Council’s actions? The council’s decision
to confine the flat rate to Atteridgeville and Mamelodi and to continue charging the metered rate in
old Pretoria was dictated by circumstances. First, the Council inherited a situation in which the
townships were not equipped with metering equipment, while houses in old Pretoria were.
Moreover, since old Pretoria is a wealthier and more developed area than Atteridgeville and
Mamelodi, it was a fair assumption that old Pretoria would have accounted for a major proportion
of the total consumption of water and electricity in the municipality. To have applied a flat rate
throughout the entire municipality would have been unscientific, and would have resulted in far
greater prejudice to individual users that the application of the flat rate in Atteridgeville and
Mamelodi alone. In the circumstances, the adoption of a flat rate as in interim arrangement while
meters were being installed in the residential areas of the two townships was the only practical
solution to the problem. As for cross-subsidisation, the Constitutional Court disagreed with the
holding of the court a quo that the levying of different rates for the same services is always unfair.
The present case was an example of an instance in which it was not unfair. The cross-subsidisation
was temporary and would be phased out once meters had been installed in the townships and a

46 De Waal, Currie & Erasmus: The Bill Of Rights Handbook 2000: 213-215.
21

consumption-based tariff introduced. In the meantime, the fact that the white town subsidised
consumption in the black townships could not be said to ‘impact adversely on the respondent in
any material way. There was no invasion of the respondent’s dignity nor was he affected in a
manner comparably serious to an invasion of his dignity.’

The selective recovery of outstanding service charges by the Council was held by the majority of
the Constitutional Court to be unfair discrimination. In old Pretoria, defaulters (such as Walker)
were summonsed and services were suspended. In the townships, there were no suspensions nor
was any legal action taken against defaulters. According to the Court, had the Council set in place
a properly-formulated policy directed at achieving the important societal goal of transforming both
the living conditions and culture of non-payment of service charges in the townships, the policy
might well have been consistent with the goal of furthering equality for all. It would therefore have
been a measure aimed at achieving substantive or restitutionary equality and not in conflict with
the equality right. Instead, the policy of taking no legal action to enforce payment of arrears had
nothing to do with the ability of residents to pay, or the introduction of metered charges and was
applicable to all residents of Atteridgeville and Mamelodi, irrespective of their financial
circumstances of their ability to pay for the services. In these circumstances, the Council has not
discharged the burden of showing that the racial discrimination was not unfair:

“No members of a racial group should be made to feel that they are not deserving of equal ‘concern,
respect and consideration’ and that the law is likely to be used against them more harshly that
others who belong to other race groups. That is the grievance that the Respondent has and it is a
grievance the council officials foresaw when they adopted their policy.The conduct of the council
officials seen as a whole over the period from June 1995 to the time of trial in May 1996 was on
the face of it discriminatory. The impact of such a policy on the respondent and other persons
similarly placed, viewed objectively in the light of the evidence on record, would in my view have
affected them in a manner which is at least comparably serious to an invasion of their dignity. This
was exacerbated by the fact that they had been misled and misinformed by the council. In the
circumstances it must be held that the presumption has not been rebutted and that the course of
conduct of which the respondent complains in this respect, amounted to unfair discrimination within
the meaning of s 8(2) of the interim Constitution.”47

5.3. Justification of unfair discrimination

5.3.1 Proportionality

Should section 9(3) or (4) be breached, it becomes necessary to determine finally whether the
measure in question can be justified in terms of the limitation clause (section 36).

This section provides that:

“(1) The rights in the Bill of Rights may be limited only in terms of law of general application
to the extent that if the limitation is reasonable and justifiable in an open and democratic

47 Par 81.
22

society based on human dignity, equality and freedom, taking into account all relevant
factors, including-

(a) the nature of the right;

(b) the importance of the purpose of the limitation;

(c) the nature and extent of the limitation;

(d) the relation between the limitation and its purpose; and

(e) less restrictive means to achieve the purpose.

(2) Except as provided in subsection (1) or in any other provision of the Constitution, no
law may limit any right entrenched in the Bill of Rights.”

Generally speaking, what has to be considered at this stage is the proportionality between the
infringement of the right to equality on the one hand and the purpose, effect and importance
of the infringing provision on the other.48 In S v Makwanyane, the court stated that the
application of the limitation clause involves a process of “weighing up of competing values, and
ultimately an assessment based on proportionality… which calls for the balancing of different
interests.”49

In balancing different interests, the relevant considerations include-

 the nature of the right that is limited and its importance to an open and democratic society
based on freedom and equality,

 the extent of the limitation of the right,

 the purpose of the limitation and the importance thereof to such a democratic society based on
freedom and equality,

 the relation of the limitation to its purpose and

 whether the limitation of the right could have been tailored in a less restrictive manner.50

These factors will be considered next.

5.3.2 Nature and importance of the right limited

48 National Coalition for Gay and Lesbian Equality v Minister of Justice and Others 1998 (12) BCLR 1517
(CC) at 1538F.
49 1995 (3) SA 391 (CC) at par 104.

50 S v Makwanyane (supra) at par 104


23

Since equality is one of the foundational values of the Constitution, a court should be extremely
cautious before upholding a justification of an act which limits the right to equality. 51 In Fraser v
Children’s Court, Pretoria North, Mahomed DP (as he then was) said that the “guarantee of equality
lies at the very heart of the Constitution [and] it permeates and defines the very ethos on which the
Constitution is premised.”52

5.3.3 Importance of the purpose of the limitation

The limitation clause requires a purpose, which is justifiable in an open and democratic society,
based on human dignity, equality and freedom for justification of the infringement of the right not
to be discriminated against unfairly.

The enforcement of the private moral views of a section of the community in respect of
homosexuality, which is based to a large extent on nothing more than prejudice, is not a purpose
important enough to justify the common law crime of sodomy, which infringes the equity rights of
homosexuals.53

The purpose of the Drugs and Drug Trafficking Act 140 of 1992 to control the use of dependence-
producing substances is an important objective to consider when deciding whether it justifiably
limits the right to freedom of religion of members of the Rastafarian religion.54

In Langemaat v Minister of Safety and Security the court rejected as a valid reason the anticipated
flood of applications and administrative difficulties that might ensue should unmarried partners
qualify as dependents for the purposes of a medical aid scheme, since it is possible, in the opinion
of the court, to consider each application on its merits.55

5.3.4 Nature and extent of the limitation

What is relevant here is the severity of the burden imposed by a limitation of the right to equality.
Relevant issues under this item may include whether a limitation substantially interferes with rights
or entails merely a minor infringement, whether it is over-broad or over-inclusive in its reach or not.

In Larbi-Odam v Member of the Executive Council for Education (N-W Province, the court found
that denying foreigners, who have been granted the right of permanent residency, permanent
appointments in teaching posts, was simply too high a price in return for increasing jobs for
citizens.56

51 Lotus River, Ottery, Grassy Park Residents Association v South Peninsula Municipality 1999 (4) BCLR
440 (C) at 452G, 454F.
52 1997 (2) BCLR 153 (CC) at 161F.

53 National Coalition for Gay and Lesbian Equality v Minister of Justice 1998 (12) BCLR 1517 (CC) at 1539G-

H.
54 Prince v President of the Law Society, Cape of Good Hope and Others 1998 (8) BCLR 976 (C) at

986E-F.
55 1998 (3) SA (T) at 449F-G.

56 1997 (12) BCLR 1655 (CC) at 1670A.


24

It was decided in Prince v President of the Law Society, Cape of Good Hope that the prohibition of
the use of cannabis undoubtedly interferes with the practice of Rastafarians of their religion, since
the use of that substance is an integral part of the practice of their religion. However, having regard
to the extent to which the eradication of the use of cannabis is in keeping with international
standards, the nature and extent of the limitation is not unreasonable.57

5.3.5 Relation between the limitation and its purpose

It is not clear which aspects of the relation between the limitation and its purpose are of interest
here. If it concerns whether there is a rational connection between the limitation and its purpose,
this factor then repeats the rational relationship test in terms of section 9(1) of the Constitution. In
Prince v President of the Law Society, Cape of Good Hope 58 the court, in applying this criterion,
found that the prohibition of the use of cannabis “advances the purposes sought to be achieved.”
This is another way of stating the rational relationship requirement. However, since the rationality
of the measure or lack thereof has already been established before or in the course of the fairness
enquiry, it does not serve any purpose to open the rational connection issue once again under the
justification analysis.

Another possibility is that the factor of the relationship of the limiting measure and its purpose is
concerned with it’s under or over-inclusiveness. If so, then especially regarding over-
inclusiveness or over-broadness, there will be an overlap with the fairness enquiry into the extent
that an infringement of the right to equality has affected the rights of the complainant, as well as
the previously discussed factor of the nature and extent of the limitation. Over-broadness or
inclusiveness is the result of a limitation that affects more categories of people than it should, or
the rights or interests of the correct category to a greater extent than it should, when assessed in
terms of what is reasonably necessary to achieve its purpose. This is the sense in which it has
been applied in Larbi-Odam v Member of the Executive Council for Education (N-W Prov).59 The
court examined the exclusion of all foreigners from permanent appointment as educators in terms
of its stated purpose, namely the reduction of unemployment. Rationally, such an aim should not
exclude permanent residents from being appointed, unless a certain post require citizenship for
some reason, for example due to the particular political sensitivity thereof.

5.3.6 Less restrictive means to achieve the purpose

In terms of this requirement, a restriction of the right to equality must not impair the right more
than is reasonably necessary to accomplish the desired objective. Therefore, if it has been
possible in the circumstances to achieve the purpose of the discrimination without discriminating
or in a less discriminating way, the discrimination will be held to be unjustifiable.

Examples

57 2002(3) BCLR 231: par 139,141.


58 1998 (8) BCLR 976 (C) at 986H.
59 1997 (12) BCLR 1655 (CC).
25

It was submitted on behalf of the applicant in Prince of the Law Society, Cape of Good Hope and
Others that the violation of the applicant’s right to freedom of religion and equality by prohibiting
the use of cannabis as part of their religious rituals, could have been avoided by a limited exemption
permitting adherents of the Rastafarian religion to possess and use cannabis for purposes of bona
fide religious observance. The court rejected this contention in the light of the fact that such an
exception would be contrary to South Africa’s obligations in terms of international conventions and
would place an additional burden on the police and the courts.60

In Brink v Kitshoff NO61, the provisions of the Insurance Act62 depriving a woman of the proceeds
of a life policy ceded to her by her deceased husband upon the sequestration of the latter’s estate,
was struck down, amongst other things because there are other non-discriminatory alternatives
available which could reasonably serve the purpose of protecting the interests of creditors against
collusion by spouses. The court referred inter alia, to provisions of the Insolvency Act in terms of
which transactions which took place prior to the insolvency could be impeached, if they were shown
to be collusive, or resulted in undue preference given to certain creditors.

6. Affirmative action

Section 9(2) states that “equality includes the full and equal enjoyment of all rights and freedoms”
and that “to promote the achievement of equality “affirmative action measures may be undertaken.”
Currie and De Waal indicate that affirmative action is not an exception to equality, but is a means
of achieving equality understood in its substantive or restitutionary sense. Apartheid created a
political and economic system that favoured some people and unfairly discriminated against others.
The right to equality does more than simply prohibit discrimination and unequal treatment by the
state or by private individuals. It also imposes a positive obligation on the government to address
the effects of previous discrimination so as to ensure that everyone fully and equally enjoys all
rights and freedoms. Affirmative action programmes must therefore be seen not as a derogation
from, but as a substantive and composite part of the right to equality. Differentiation aimed at
protecting or advancing persons disadvantaged by unfair discrimination is therefore warranted
provided the measures are shown to conform to the internal test of section 9(2). Practically this
means that when a measure has been challenged as a violation of the equality right, the state or
the institution responsible for the measure can defend it by showing that the measure (1) targets
persons or categories of persons who have been disadvantaged by unfair discrimination; (2) is
designed to protect and advance such persons or categories of persons and (3) promotes the
achievement of equality.63

Affirmative action will be separately discussed in more detail in unit 4 after discussion of the
Employment Equity Act and the Promotion of Equality and Prevention of Unfair Discriminaton Act

60 2002(3) BCLR 231: par 139,141.


61 1996 (6) BCLR 752 (CC) at 770J-771C.
62 Section 44 of Act 27 of 1943.

63 Currie & De Waal. The Bill of Rights Handbook 2005: 264-265.


26

which contain affirmative action provisions. Examples of the implementation of affirmative action
especially in the employment context will serve to illustrate its application.
27

UNIT 2

GENERAL PRINCIPLES OF EMPLOYMENT DISCRIMINATION LAW

A Introduction
Section 9(4) of the Constitution makes provision for national legislation to prevent or prohibit unfair
discrimination. The constitution also states, in section 9(2), that legislative and other measures
must be taken to promote the equal enjoyment of rights of persons or categories of persons
disadvantaged by discrimination.

The first piece of legislation to give effect to the Constitution in this respect is the Employment
Equity Act 55 of 1998 (EEA). The Act places two main obligations on employers:

 The prohibition of unfair discrimination in the workplace (chapter II);

 The adoption of affirmative action (chapter III)

All employers are obliged to prevent and eliminate unfair discrimination in the workplace. Apart
from the eradication and prevention of unfair discrimination in the workplace, it also tries to ensure
that employers take positive measures to transform their organisations and remove barriers to the
equitable treatment of previously disadvantaged groups. The rationale for the implementation of
employment equity centres around eradicating the legacy of discrimination in respect of race,
gender and disability that has denied South Africans the opportunity for education, employment,
promotion and wealth creation. (Tinarelli Employers’ Guide to the Employment Equity Act p.15).

While all employers are obliged to eliminate unfair discrimination in the workplace, only so-called
designated employers have to implement affirmative action in favour of so-called designated
groups.

Designated employers:

Employers who employ more that 50 people or who have an annual turnover of between R 2-
million and R 25-million, depending on the industry, are considered “designated employers” and
are required to abide by the provisions of chapter 3 of the Act.

Designated groups:

Designated groups include blacks (i.e. Africans, Indians and Coloureds), women and people with
disabilities.

In this section we will first discuss the general principles of employment equity and then focus on
discrimination in relation to specific employment practices and policies as well as harassment.
Thereafter, the thorny issue of affirmative action will be addressed.
28

B Prohibition of employment discrimination

1 Provisions of the EEA


Section 5 of the EEA:

Every employer must take steps to promote equal employment in the workplace by
eliminating unfair discrimination in any employment policy or practice.

Section 6 of the EEA:

(1) No person may unfairly discriminate, directly or indirectly, against an employee, in


any employment policy or practice, on one or more grounds, including race, gender,
sex, pregnancy, marital status, family responsibility, ethnic or social origin, colour,
sexual orientation, age, disability, religion, HIV status, conscience, belief, political
opinion, culture, language and birth.

(2) It is not unfair discrimination to —

(a) Take affirmative action measure consistent with the purposes of this Act, or

(b) Distinguish, exclude or prefer any person on the basis of an inherent requirement
of a job.

(3) Harassment of an employee is a form of unfair discrimination and is prohibited on


any one, or a combination of grounds of unfair discrimination listed in subsection
(1).

Section 1 of the EEA

defines “employment policy or practice” as including the following: recruitment procedures,


advertising and selection criteria; appointments and the appointment process; job classification
and grading; remuneration, employment benefits and terms and conditions of employment; job
assignments; working environment and facilities; training and development; performance
evaluation systems; promotion; transfer; demotion; disciplinary measures other than dismissal;
and dismissal.64

Section 11 of the EEA:

Whenever unfair discrimination is alleged in terms of this Act, the employer against whom
the allegation is made must establish that it is fair.

2 Direct and indirect employment discrimination

64 Section 1. See also section 14 of the Promotion of Equality and Prevention of Unfair Discrimination Act 4
of 2000, which lists some of the same categories of employment practices to which the prohibition of unfair
discrimination applies
29

Section 6 of the EEA prohibits direct as well as indirect discrimination in employment.

Direct discrimination

Direct discrimination refers to an act of discrimination based on a prohibited ground. In terms of


the constitutional scheme, direct discrimination involves discrimination based on specified as well
as analogous non-specified grounds, since both are prohibited grounds of discrimination. Non-
specified prohibited grounds of discrimination concern distinctions based on attributes or
characteristics that have the potential to impair the fundamental dignity of persons as human
beings or to affect them adversely in a comparably serious manner.

An employment policy or practice discriminating against non-citizens, for example, will be directly
discriminatory, although citizenship as such is not mentioned specifically as a ground of
discrimination in section 6 of the Employment Equity Act. A differentiation on account of citizenship
has been found to constitute a distinction based on an attribute with the potential to impair human
dignity in the Larbi-Odam case.65

Direct unfair employment discrimination, in terms of the scheme developed by the Constitutional
Court, will thus occur when persons are disadvantaged in any employment practice, policy or
procedure on any ground specified in section 6 of the Employment Equity Act, or on other
unspecified attributes or characteristics, which either have the potential to impair their fundamental
dignity as human beings, or affect them adversely in a comparably serious manner.

Indirect discrimination

Indirect discrimination occurs when an employment policy or practice is based on a non-prohibited


ground (e.g. a person’s appearance or work experience), but its effect is that a category or group
of persons, protected under a prohibited ground of discrimination, is adversely affected. For
instance, deciding not to employ somebody because he or she is under a specified height, does
not fall directly under any of the specified or analogous grounds, but may in specific circumstances
indirectly affect one of the specified grounds (e.g. gender, ethnicity or race), or may well constitute
a form of discrimination that indirectly impinges on the dignity of the individual. 66

3 Proof of direct employment discrimination


When does an act or omission constitute direct discrimination in the workplace? In this section,
the elements of direct employment discrimination will be discussed. Remember that the
Constitutional Court has decided that the question whether an act or omission constitutes unfair
discrimination involves two steps:

65 Larbi-Odam and Others v Member of the Executive Council for Education and Another (N-W Province)
1997 (12) BCLR 1655 (CC).
66 In given circumstances, discriminating in terms of physical characteristics may also be direct discrimination

based on characteristics with the potential to impair human dignity or to affect the complainant in a
comparably serious way.
30

 Firstly, is there “discrimination”, and

 Secondly, is the discrimination unfair?

It is only the first question that we will be addressing here. The issue of fairness will be discussed
infra .We are addressing the first step in this paragraph and paragraph 4. The issue of fairness
(justification) will be discussed in paragraph 5 infra.

How does a complainant prove (prima facie) direct discrimination?

The complainant must prove that the prohibited ground is the reason or basis of the differential
treatment.67 The Act prohibits unfair distinctions based on certain specified and analogous
grounds, such as someone’s race, gender, religion, marital status, etc., or personal characteristics
that could infringe someone’s dignity or in a comparably serious way affect him or her. Distinctions
based on other grounds are not prohibited. It is therefore up to a complainant to prove that he or
she has been treated adversely because of his or her race, gender, religion, etc.

In order to establish that a prohibited ground is the actual basis of a discriminatory act, it is
necessary to make certain that the circumstances of the groups or categories compared are
substantially so similar that they ought to be treated alike, or so dissimilar that they ought to be
treated differently. The mere fact of different treatment is therefore not enough to establish
discrimination.68 For instance, a difference in salary paid to a male and female employee as such
does not amount to prima facie discrimination. It will only do so in situations in which the relevant
circumstances of the employees are so comparable that they ought to have been paid the same.

Example

In Louw v Golden Arrow Bus Services (Pty) Ltd69, the applicant, a black male, was employed as a
security guard and was offered a “marketing job” together with two other former shop stewards.
He complained about the fact that they were paid salaries and commissions substantially lower
than that received by a white male, who was employed as a warehouse supervisor. The Labour
Court held that discrimination takes place when similarly circumstanced individuals are treated
differently. In order to make out a prima facie case of race discrimination, it must be established
that the respondent has treated similarly situated white persons differently. The court held
that in the circumstances of the case, the situations of the applicant and the white comparator were
materially different, in the light of the different nature of the particular jobs. The jobs involved
different levels of responsibility, expertise, experience, and skills. It found that the applicant failed
to show that white and black employees that are similarly situated in relation to these factors were
being treated differently.

67 Cf Louw v Golden Arrow Bus Services (Pty) Ltd (1999) 8 LC 6.12.5 at par 26.
68 See also Swanepoel v Western Region District Council and Another [1998] 9 BLLR 987 (SE) at 990-H;
TGWU v Bayete Holdings [1999] 4 BLLR 401 (LC).
69 (1999) 8 LC 6.12.5 (case no J2521/98).
31

It is also important to note that section 6 of the EEA prohibits discrimination on “one or more” of
the specified or analogous prohibited grounds. This means that a complainant need not prove
that the alleged discrimination is solely on the basis of any one of the prohibited grounds.
A claim may be based on a combination of prohibited grounds.70 This has special significance in
the light of the substantive equality approach of the Constitutional Court in terms of which the
fairness of a discriminatory act is measured primarily with reference to its impact on the victim of
discrimination, taking into account the position of the complainants in society and the nature and
extent of the negative effect of the measure on their rights or interests. The fact that complainants
are not confined to present their claims as representatives of separate generic categories only
makes it possible for them to demonstrate the unique ways in which discriminatory treatment is
experienced by groups defined by a combination of attributes, such as black or married women,
older disabled persons, HIV-positive young children etc.

Illustration

The practical importance of recognising claims based on a combination of grounds of discrimination


can be illustrated by the American case of Arnett v Aspin.71 The plaintiff, a woman of over forty
had been passed over for promotion twice. Her case of sex-plus-age discrimination was based on
the fact that all women who had been promoted were younger than forty and every man promoted
was over forty. If she had to present her case as either sex or age discrimination, as the employer
contended, she would have been faced with the difficulty that both younger women and older men
had been selected for positions in question. In casu the court rejected the employer’s contention
and held

“[f]or purposes of determining whether the defendants discriminated against Arnett in


violation of Title VII, I find she is a member of a discrete subclass of ‘women over forty’.
Accordingly, I conclude that Arnett has shown a prima facie case … because (1) she is a member
of the protected subclass, that is women over forty, (2) she was qualified for and applied for the
positions in question, (3) despite her qualification, she was denied the positions, and (4) other
employees outside her protected class were selected, in this case two women under forty.

It is important to note that an employer’s reasons or motives for discriminating are irrelevant
for the question whether a prima facie case of discrimination has been established. The
constitutional framework for establishing unfair discrimination presupposes a clear distinction
between the basis or ground of differential treatment and the reasons or purposes advanced in
justification thereof. In satisfying the first leg of the unfair discrimination enquiry, namely whether
employment conduct amounts to “discrimination”, all that is required is that an employment practice

70 See Brink v Kitshoff [1996] 5 BCLR 752 (CC) at par 43; President of the Republic of South Africa v Hugo
[1997] 6 BCLR 708 (CC) at par 33.
71 846 F Supp 1234 (ED Pa 1994).
32

or policy distinguishes between persons in terms of a prohibited ground, either directly or indirectly.
Only when this is established, will the issue of fairness be raised.72

For the first leg of the enquiry, it is immaterial what the objective or subjective reasons for, or
purposes with the differential treatment were. An employer’s motivating reasons or concerns are,
however, important considerations in rebutting the presumption of fairness or justifying
discriminatory actions in terms of the limitation clause of the Constitution.73 Maintaining the
distinction between causation (on what criterion is the differential treatment based?) and
justification (what was the employer’s reason for or purpose with the differential treatment?), is thus
an important element of the analytical process of establishing unfair discrimination.74

Illustration

In Peake v Automotive Products Ltd75, the court had to decide whether the decision of an employer
to allow female employees to leave work five minutes earlier than men, was discrimination based
on sex. The employer maintained that the difference in treatment was for purposes of
administrative efficiency and to prevent the female employees from being jostled and hurt in the
crush at the factory gates. In its rejection of the complainants’ contention that they had been directly
discriminated against, the Court concluded that the grounds for the differential treatment were
administrative efficiency, safety and traditional courtesy, rather than sex. Lord Denning MR was
of the opinion that it would be wrong to “obliterate the differences between men and women or to
do away with the chivalry and courtesy which we expect mankind to give to womankind.”76 In his
opinion, therefore, this was not a case of discrimination based on sex, but rather on the admirable
motive of treating women in a chivalrous manner.77 In the later decision of Jeremiah v Ministry of
Defence78, the Court of Appeal reversed its approach in Peake. Lord Denning MR then decided
that the Court of Appeal in Peake had erred in finding that, because of the motivating concerns of
administrative efficiency and chivalry, the differentiating treatment was not based on sex. In the
Jeremiah case, only male workers were required to work in an unpleasant and dirty part of the
employer’s factory. Women were excused from these tasks, because the employer felt that the
work was too offensive and that there was insufficient showering facilities for women. The Court of
Appeal held that the employer had treated the male employees less favourably than their female
counterparts and that this treatment was based on the sex of the complainants. The fact that the
employer was not motivated by any intent to do the complainants any harm or that its moves were
benign, did not change the matter.

72 Harksen v Lane NO and Others 1997 (11) BCLR 1489 (CC) at 1508A-D.
73 S 36.
74 “Grounds” and “reasons” may, in specific circumstances, coincide, which will mostly be the case with

intentional direct discrimination.


75 [1977] ICR 968 (CA).

76 Id at 973B-C.

77 Peyton Sex and race discrimination law (1997) 18.

78 [1979] IRLR 436 (CA).


33

This case illustrates why it is important to distinguish the “ground” in terms of which a measure
differentiates from the “purposes” or “reasons” for the differentiation. If it is not done, behaviour that
constitutes prima facie “discrimination” might not be identified as such and no reason would exist
for the employer to justify the prima facie discriminatory action.

From the discussion of the Constitutional Court’s jurisprudence, it appears that intention to
discriminate as such is not a requirement for unfair discrimination. The question is simply whether
an employee suffers adverse treatment based on a prohibited ground. The presence or absence
of the intention to discriminate does not affect the direct nature of the discrimination as such.

4 Indirect employment discrimination


4.1 Introduction

We have defined indirect employment discrimination in the previous section as follows:

Indirect discrimination occurs when an employment policy or practice is based on a non


prohibited ground (e.g. a person’s qualifications or work experience), but its effect is that a
category or group of persons, protected under a prohibited ground of discrimination, is
adversely affected (e.g. a gender, religious or ethnic group).

Example

The Labour Court in Leonard Dingler Employee Representative Council v Leonard Dingler (Pty)
Ltd79 explained that indirect race discrimination occurs when criteria, conditions or policies are
applied that appear to be neutral, but which adversely affect a disproportionate number of a certain
group in circumstances where they are not justifiable. In this case, the apparently neutral practice
of paying employees on a monthly or weekly basis was found to have had an indirectly
discriminatory effect on a racial basis. Only monthly paid employees, who were mainly white,
qualified for the staff benefit fund. Most black employees were paid on a weekly basis, which
meant that they did not qualify to participate in the pension fund.

4.2 Locus classicus

It is the United States Supreme Court which is usually credited with the first unequivocal recognition
of the concept of indirect discrimination in employment. In Griggs v Duke Power Co80 the Court
held that Title VII of the Civil Rights Act 1964, in addition to outlawing intentional discrimination,
also prohibits neutral employment practices which have a “disparate impact” upon protected
groups of employees, unless those practices are shown to be justified by “business necessity”. The
case concerned the company’s practice of requiring a high school education or passing of a general
standardised intelligence test as condition of employment in or transfer to certain jobs. The

79 (1997) 2 LC 6.12.1 at 6

80 401 US 424, 91 SCt 849 (1971)


34

requirements had the result that black employees were disqualified at a higher rate than whites.81
The company failed to demonstrate that either of the two requirements reliably measured future
job performance, since employees who had not met either requirement continued to perform and
progress satisfactorily in the particular departments. In fact, both were introduced without any prior
meaningful study of their relationship to job performance ability. The Court held that neutral
employment procedures or testing mechanisms that are unrelated to job performance and which
operate as “built in headwinds” for many minority groups fall within the ambit of unlawful
employment practices under Title VII of the Civil Rights Act 1964. The Act does not command that
any person be hired merely because he was formerly the subject of discrimination, or because he
is a member of a minority group. Discriminatory preference for any group, minority or majority, is
precisely what the Act proscribes. What is required is the removal of artificial, arbitrary, and
unnecessary barriers to employment when the barriers operate invidiously to discriminate on the
basis or racial or other impermissible classification. The Court concluded that-

“Congress has now provided that tests or criteria for employment or promotion may not
provide equality of opportunity merely in the sense of the fabled offer of milk to the stork
and the fox. On the contrary, Congress has now required that the posture and condition of
the job-seeker be taken into account. It has - to resort again to fable – provided that the
vessel in which the milk is proffered be one all seekers can use. The Act proscribes not
only over discrimination but also practices that are fair in form, but discriminatory in
operation. The touchstone is business necessity. If an employment practice which
operates to exclude Negroes cannot be shown to be related to job performance, the
practice is prohibited.”

Griggs thus established the principle that when an employer relies on recruitment policies or testing
devices which have an adverse effect on protected groups, a burden is imposed on the employer
to justify the policy. If an employer fails to satisfy the test of “business necessity” and “job
relatedness” the practice will be declared unlawful.

4.3 Elements of indirect employment discrimination

A complainant must prove that-

 a neutral employment policy or practice

 causes detriment in a disproportional manner

 to a category of people protected under a prohibited ground of discrimination.

(a) Neutral employment policy or practice as the cause of detriment to a category of people
protected under a prohibited ground (causality)

81 The Court (91 SCt 849 at 853 (1971)) noted that “[b]ecause they are Negroes, petitioners have long
received inferior education in segregated schools…”.
35

Firstly, a complainant has to prove that a particular employment policy or practice has a
discriminating effect.

S. 1 of the Employment Equity Act defines “employment policies and practices” to include – but not
to be limited to – the following: recruitment procedures, advertising and selection criteria,
appointments and the appointment process, job classification and grading, remuneration,
employment benefits, terms and conditions of employment, job assignments, working environment
and facilities, training and development, performance and evaluation systems, promotion, transfer,
demotion, disciplinary measures other than dismissal, and dismissal.

The employment policy or practice must be neutral, i.e. it must be one that applies to all employees
concerned, irrespective of their membership of a class or group protected in terms of the prohibited
grounds of discrimination. The nature of indirect discrimination is such that the employment policy
or practice should be one that discriminates in its effects only.

Secondly, complainants will have to prove that it is the particular employment policy or practice
that causes the alleged disparity.

Illustration

In Home Office v Holmes82, a full-time female employee with two young children applied to work
part-time. The employer refused to grant the request because it took the inflexible position that
people employed in her specific job category had to work full-time. The tribunal took the view that
this policy discriminated indirectly on the basis of sex. Because the responsibility of taking care of
children is still primarily borne by women, it is much more difficult for women with family
commitments than for men to be able to work full-time. Since the employer could not show why the
nature for the job required full-time employment, he failed to justify the indirectly discriminatory
policy.

In Hill v Seaboard Coast Line RR Co83 the difference in promotion rates between blacks and whites
was found to be the result of a lack of interest on the part of the black group and not because there
was something wrong with the promotion process as such. Although the pool from which
promotions were done was 12% black, they made up only 4% of those promoted. No prima facie
indirect discrimination was found because many blacks that were offered promotions turned it
down.

The complainant need not prove that the employment practice or policy is the only or the main
reason for the disadvantage suffered by the complainant group. It is enough if the policy or practice
is a factor in the decision making process. Therefore, if an employment practice, such as a
requirement for selection or promotion, disproportionally disadvantages protected groups, but it is
only one of a number of factors taken into account, it still qualifies as a discriminatory practice. Our

82 [1984] ICR 678.


83 885 F2d 804 (11th Cir 1989).
36

courts should therefore not follow the approach taken in the case of Perera v Civil Service
Commission.84 The policy complained of was the requirement of experience as a solicitor in
England and fluency in English to be appointed as a legal assistant in the civil service. The court
rejected the claim that the requirements indirectly discriminated against immigrants. Stephenson
LJ held that a person claiming to have been indirectly discriminated against must establish that the
allegedly discriminatory requirement was “a ‘must’; something which has to be complied with.”
Since the allegedly discriminatory requirements were but some of the factors taken into account,
the lack of it was not an absolute bar to appointment. In the court’s opinion, it was always possible
that the complainant’s performance regarding the other requirements could have compensated for
a poor performance in respect of the allegedly discriminatory factors.

The “absolute bar approach” has correctly been criticised as unduly restrictive and ought no to be
followed when applying the Employment Equity Act. If a factor or a collection of factors, while
applied to all, tends to operate unfavourably against a particular class of people, such factors ought
in the circumstances of that particular case to be seen as a discriminatory requirements or
conditions. In the competitive modern labour market, any criterion that considerably disadvantages
an applicant for employment is in effect a bar to that candidate’s success.

(b) Disproportional impact

The central element of indirect discrimination involves the requirement that an employment policy
or practice detrimentally affects a substantially higher proportion of members of a protected
group than members of a comparable group.

The process for establishing whether an employment policy or practice has such a disproportional
adverse effect on a specific group, involves three distinct analytical steps:

 firstly, identifying the correct pool of comparison;

 secondly, establishing whether a practice affects the compared groups in the identified pool
disproportionally; and if it does,

 thirdly, determining whether the disproportionality is substantial.

(b.1) Identifying the pool of comparison

In order to determine whether a practice affects one group more disadvantageously than another,
the group of affected persons that will form the basis of the comparison (the sample group) must
be identified accurately. Once a pool is selected, the persons in the pool of the complainant’s group
provide the numbers for one side of the proportion calculation, and the persons in the pool of the
comparator group provide the numbers for the other side of the proportion calculation.

For example

84 [1983] IRLR 166 (CA)


37

A company has the policy that only monthly-paid employees qualify for the company’s pension
fund. The total workforce of the company is 100, of which 60 are black and 40 are white. Ninety
percent of black employees are paid on a weekly basis, while ninety-five percent of white
employees are paid on a monthly basis. The result of this apparently race-neutral policy is
therefore that most of the black employees are excluded from the pension fund. In order to prove
indirect race discrimination, the complainants have to show that proportionally more blacks than
whites are disadvantaged by the policy. The sample group or pool of comparison in this case is
all the employees of the company. In this sample group, 95% of the white portion of the group are
monthly paid and qualify for the pension, while only 10% of the black portion of the sample group
qualifies.

Very often it will be a matter of common knowledge that certain policies or practices disadvantage
certain groups. No elaborate statistical evidence will be necessary to prove indirect discrimination
in this case and courts can take judicial notice thereof. This will be the case with employment
practices that inherently harbour elements of systemic or structural discrimination.
Examples of such practices are formal qualifications and work experience requirements, because
of our history of educational discrimination and deprivation, as well as the practice of excluding
blacks and women from certain jobs. A good example of how such policies or practices should be
approached is the case of London Underground Ltd v Edwards (No 2).85 The complainant
challenged the employer’s shift system, alleging indirect sex discrimination because of women’s
childcare responsibilities. All full-time employees were required to work certain night shifts. The
complainant was only one out of 21 female employees who could not comply with the requirement,
a result suggesting little comparative difference in the adverse effect on women and men as a class
in the employer’s workforce. If the comparison had been restricted to the specific workforce, the
evidence would thus not have shown a disproportionate sex-related effect. However, the tribunal
correctly held that the requirement did in fact indirectly discriminate because of sex. Morrison J
stated that the tribunal could take judicial notice of the fact that generally proportionally a larger
number of women than men in employment have childcare responsibilities. That this general
picture is not mirrored in the specific workforce of the company, is a matter of co-incidence and
does not detract from the fact that such policies prejudice women more than men

(b.2) Establishing the relevant proportions

Once the correct pool has been identified, the next step is to determine the proportions of the
relevant groups within the pool that are detrimentally affected by the policy or practice in question.
If, for instance, the complaint concerns indirect sex discrimination because of a requirement of full-
time service to be eligible for promotion, what has to be established is what percentage of the men
and what percentage of the women in the relevant pool are fulltime workers.

85 [1997] IRLR 157 (EAT)


38

It is the proportion (percentage) of each group and not total numbers that has to be compared.
Indirect discrimination is not proved by merely comparing the number of persons from one group
to the number of persons of another group in a given pool. Numbers as such may, for instance,
indicate the gender or racial balance in a particular workforce or labour market but do not
necessarily provide evidence of a disproportional impact of a policy or practice on protected groups.

For example

An employer has a vacancy for a typist. The sample group or base of comparison is all eligible
typists in the relevant labour market (i.e. the pool of people from which the employer may
reasonably be expected to recruit). In the city in which his business is located, there are 200
qualified typists, 80% of whom are female. In its advertisement, the employer specifies as a
minimum requirement that the candidate must be 2m tall. 4 of the 160 females of the sample group
(i.e. 2.5%) comply, whereas 4 of the 40 male candidates in the sample group (i.e. 10%) comply.
Although the same number of male and female candidates complies, the percentages show that
the requirement disproportionally favours men.

(b.3) Determining substantial disproportionality

The third step in the disproportional impact analysis concerns the size of the disparity: how
disproportional must the effect of an employment policy or practice be in order to be found
indirectly discriminatory? Must a practice or policy have a greatly disproportional impact, or is any
deviation in the way it affects different groups sufficient?

Although there is no fixed standard, the common approach in other jurisdictions seems to be that
the disparity must be substantial, but differences in interpretation and practical application
obviously occur.86 It seems that, in the normal course of events, the threshold standard of a
substantial disparity in the way the practice affects different groups is – from an evidentiary point
of view – necessary to establish that the disparity is actually group-based. Small differences may
not persuasively point to group-related disadvantage or rule out statistical coincidences.

5 Justification of employment discrimination


5.1 Introduction

This section considers the possible justifications that an employer may raise as a defence to a
claim of unfair discrimination.

Section 6 of the EEA mentions only two defences. It states that it is not unfair discrimination to-

86In the US, the agency administering Title VII of the Civil Rights Act 1964, the Equal Employment
Opportunity Commission has put forward a “four-fifths” rule of thumb, whereby any difference between
compliance rates of more than 20% is “substantial”
39

 Take affirmative action measures consistent with the purposes of the Act, or

 To distinguish, exclude or prefer any person on the basis of an inherent requirement of a job.

Employer defences, however, are not restricted to the two expressly mentioned in the Act. From
the analysis of the discrimination jurisprudence of the Constitutional Court, it appears that the
burden of justification may arise in two contexts, namely in respect of establishing that the unequal
treatment in employment is not unfair and in the context of justifying unfair discrimination. The wide
range of relevant factors for the inquiry into unfairness and justification makes clear that more
considerations than inherent job requirements or affirmative action may justify differential employer
conduct.

The grounds that employers may legitimately rely on in justification of discriminatory treatment can
be classified under three main headings:

 Justification based on the inherent requirements of the job. This ground of justification
relates to distinctions necessary for the performance of the actual duties of the job, for example
the requirement that an employee should possess specific qualifications essential for the
effective carrying out of the duties for the job.

 The operational requirements of the enterprise or undertaking. This concerns grounds of


justification which are related to the general effective running of the employer’s enterprise as a
whole, for example, the requirements that employees work full-time, as otherwise business will
suffer under the cost burden of extra changes of shifts necessitated by part-time employees,
and

 Public interest-related justifications. A public interest-related ground of justification is one


that is related to the broad public or social benefit provided by a differential requirement or
condition, for example the granting of special social assistance to those individuals or groups
with the greatest need; and

 Affirmative action

The specific categories of employer justifications will be discussed in detail, with reference to case
law examples drawn from South African and other jurisdictions. The affirmative action defence
against a claim of employment discrimination will be discussed later in a separate part.

5.2 Inherent requirements of the job

(a) Meaning of “inherent job requirements”

The term “inherent” implies a job requirements that is part of the “essential features or defining
characteristics” of the position in question.
40

In Whitehead v Woolworhs (Pty) Ltd87, the Labour Court dealt with the question whether
uninterrupted job continuity was an inherent requirement for the position of a human resources
officer. The respondent declined to offer the job to the complainant because of the interruption in
service that her pregnancy would have brought about. Waglay J held that an inherent requirement
implies an “indispensable attribute” of the job. An inherent requirement is one that if not met an
applicant would simply not qualify for the post. If the job can be performed without the requirements,
it cannot be said that it is inherent. Applied to the fact of the case, this meant that uninterrupted job
continuity could only be considered an inherent job requirement if time was of the essence, such
as when there is some contract that needed to be performed within a prescribed period. Since this
was not the case, the Court found in favour of the applicant.

Job qualifications that are “peripheral to the central mission of the employer’s business”, cannot be
considered as inherent job requirements. The job requirements, which an employer invokes to
justify its discrimination, must be reasonably necessary to the “essence” of the business.

In Diaz v Pan American88, the court rejected the notion that being female was an inherent job
requirement for an in-flight cabin attendant. The airline contended that passengers preferred the
“pleasant environment” and the “cosmetic effect” provided by female attendants, and that most
men were unable to perform these “non-mechanical” functions of the job effectively. The court
concluded that these considerations were “tangential to the essence of the business involved”,
namely the transport of passengers.

Most other jurisdictions accept that the concept of inherent job requirements may include more
than the specific task or function implicit in the job description, to include also the circumstances
under which the job may be safely and effectively performed. The concept of inherent job
requirements is, therefore, wide enough to include inter alia also the following:

 The ability to perform the job safely and not to pose a health or safety risk to others,

 The ability to perform the job in a cost-effective and profitable way,

 The ability to comply with the legitimate preferences of customers or other third parties (such
as legitimate privacy or modesty preferences, but excluding preferences that are discriminatory
in themselves).

(b) Health and safety

Courts in other jurisdictions have concluded that it is permissible to take into account the health
and safety of the complainant, co-workers or others, when considering the requirements of
employment.

Examples

87 (1999) 8 LC 6.12.4.
88 442 F2d 385 (5th Cir 1971)
41

In X v The Commonwealth89, the appellant was discharged from the Australian army because of
his HIV positive status. The respondent alleged that the appellant was unable to carry out the
inherent requirements of the particular employment. This contention was not based on the
appellant’s physical incapability to carry out combat-related tasks as a soldier, but on the risk he
might pose to other soldiers by reason of his HIV infection. It was submitted that the risk of
transmission of HIV infection to other soldiers means that an HIV infected soldier is not a suitable
candidate for deployment and that because of the infection, the appellant would be unable to carry
out an inherent requirement of his employment, namely deployment as required. Although the
court remitted the question whether the appellant’s condition in fact posed a serious safety risk to
other back to the Commission, it addressed the issue to what extent safety considerations may
qualify as inherent job requirements. The Court rejected the argument that because the appellant
was physically able to perform the tasks involved in actual battle, his dismissal could not be justified
on the basis that he could not perform an inherent job requirement. The majority accepted that
inherent requirements refer to the “characteristic or essential requirements of the employment as
opposed to those requirements that might be described a peripheral”. However, inherent job
requirements do not only refer to what the employee is to do or to be trained for, but also the
circumstances in which the particular job will be performed. Those circumstances will often include
the place or places at which the employment is to be performed and may also encompass other
considerations. For example, it may be necessary to consider the dangers to which the employee
may expose himself and dangers to which the employee may expose others. Confining attention
to the tasks and skills for which a soldier is specifically prepared was too narrow a focus in the
present case. It left out the consideration where, when, in what circumstances, and with whom
those tasks and skills were to be performed or used. All of these features form part of the inherent
requirements of the job.

The American Supreme Court recognised sex as an inherent job requirement in circumstances
where the employment of a female guard in contact areas of maximum security male penitentiaries
would create real risks to others if violence broke out because the guard was a woman. Sex
discrimination was justified because sex was related to the guard’s ability to do the job, namely
maintaining prison security.90

In Western Airlines v Crisswell91, the court considered safety to third parties, in respect of a
complaint regarding the airlines’ policy of a compulsory retirement age of sixty for flight engineers.
The court focused on the nature of the flight engineer’s tasks and the actual capabilities of persons
over 60 in relation to those tasks. The retirement policy was held to be lawful, because of age-
connected debility, a flight engineer might not properly assist the pilot, and might thereby cause a
safety emergency.

89 [1999] HCA 63
90 Dothard v Rawlinson 433 US 321 (1977)
91 472 US 400 at 406
42

It is not enough that a person poses a slight risk to third parties. The factors that must be
taken into account to decide whether a safety risk is substantial enough to justify the
exclusion of persons from job opportunities in terms of a prohibited ground are the following:

 Degree of risk (in the sense of the probability that it will be realised) and

 The consequences of it being realised (in the sense of the seriousness of the harm that will
ensue if it is realised).

The greater the safety factor, measured by the likelihood of harm and the probable severity of that
harm in case of an accident, the more stringent may the job qualification designed to insure safety
be.

In School Board of Nassau County v Arline92, the court dealt with the dismissal of a schoolteacher
because of her susceptibility to tuberculosis. The court held that regard should be had to, inter alia,
the nature and the duration of the risk of transmission of tuberculosis, the probability of
transmission, and the severity of the consequence of the transmission.

“The court concludes that plaintiff posed no threat of communicating tuberculosis to the
school children she was teaching ... the probability that she would transmit tuberculosis to anyone
was so extremely small as to not exist. Plaintiff’s smear tests were negative, the positive 3culture
of November 1978 had only one colony and was surrounded by negative cultures taken the same
day and in the adjoining months, plaintiff was on medication. None of the family tested positive for
tuberculosis, and the exposure to the students was of a limited time. All experts who testified at
the trial … stated that plaintiff is cured of tuberculosis.”

In Hoffmann v South African Airways93 Mr Hoffmann challenged the constitutionality of the South
African Airways’ policy of not employing people living with HIV as cabin attendants in the High
Court. SAA defended its policy as promoting the safety and health of its passengers and its own
competitive capacity: people living with HIV might not react positively to yellow fever vaccination,
they might contract and transmit other diseases to passengers and are a bad training investment
because of their limited life expectancy. The High Court upheld SAA’s defence. It found that the
practice: was “based on considerations of medical, safety and operational grounds”; did not
exclude persons with HIV from employment in all positions within SAA, but only from cabin crew
positions; and was “aimed at achieving a worthy and important societal goal.” The High Court
furthermore noted that if the employment practices of SAA were not seen to promote the health
and safety of its passengers and crew, its “commercial operation, and therefore the public
perception about it, will be seriously impaired”. A further factor that it took into consideration was
the allegation by SAA that its competitors apply a similar employment policy. The court reasoned
that if SAA were obliged to employ people with HIV, it “would be seriously disadvantaged as against
its competitors”. It concluded that “it is an inherent requirement for a flight attendant, at least for

92 480 US 273 at 288, 107 SCT 1123 at 1132, 94 Led 2d 307 (1987)
93 Hoffmann v South African Airways (CCT17/00) [2000] ZACC 17; 2001 (1) SA 1; 2000 (11) BCLR 1235.
43

the moment, to be HIV-negative” and that the practice did not unfairly discriminate against persons
who are HIV positive.

In a unanimous decision written by Justice Ngcobo, the Constitutional Court held that SAA had
infringed Mr Hoffmann’s constitutional right not to be unfairly discriminated against. Based on
medical evidence, the Court concluded that not every person who is HIV positive poses the risks
alleged by SAA, only those who are at the immunosuppressed stage. Mr Hoffmann was not
immunosuppressed either at the time he applied for the position or when he brought the matter to
the Constitutional Court.

The Court held that legitimate commercial requirements are important but they cannot serve to
disguise stereotyping and prejudice, which have no place in this era of respect for human dignity,
compassion and understanding - ubuntu. People living with HIV have been stigmatised and as one
of the most disadvantaged groups in society deserve special protection from our law.

(c) Cost and profitability

Accommodating previously excluded classes may pose initial direct cost to an employer. For
instance, appointing women in occupations from which they were previously excluded may bring
about costs such as the construction of bathroom facilities, the purchase of new and different styles
of uniforms, installation of lighting to deter sexual assault, etc. The absence of an employee who
becomes pregnant inevitably gives rise to cost for the employer. Terminating patterns of
discrimination may also cause some indirect cost through lost efficiency flowing from disruptions
of employee morale or short-term customer dissatisfaction.

Since the emphasis of the inherent requirement defence is on the ability of the employee to perform
job-related tasks, cost and profitability considerations as such cannot constitute the basis
of an inherent job requirement defence for the exclusion of groups or individuals concerned.
The provision of separate sanitary facilities, for instance, is a condition regarding the organisation
of the working environment and not something related to the nature of the job activity itself.

In Smallwood v United Airlines, Inc94, for instance, the court rejected the contention that any
practice that enables an employer to save costs and maximise profit should be considered an
essential element of the employers’ business. The airline tried to justify its rule of not offering
employment to pilots over 35 years of age with reference to the high cost involved in training and
the fact that appointing only young pilots allowed a longer period of utilisation of employees.

In Whitehead v Woolworths (Pty) Ltd95, the Court rejected the argument that the exclusion of the
complainant from employment due to pregnancy could be justified because of commercial
concerns. He argued that “if profitability is to dictate whether or not discrimination is unfair it would
negate the very essence for the need of a Bill of Rights.”

94 661 F2d 303 (4th Cir 1981)


95 (1999) 8 LC 6.12.14 Para 27-30
44

Also see Hoffmann v South African Airways (CCT17/00) [2000] ZACC 17; 2001 (1) SA 1; 2000
(11) BCLR 1235

It is possible however, that some cost or financial risks go beyond the foreseeable, short term,
incidental expenses of engaging in non- discriminatory employment policies.96 It has been held, in
the Johnson Controls97 cases that cost may be a defence in a situation where cost “would be so
prohibitive as to threaten the survival of the employers’ business” 98 The Canadian Human Rights
Act 1985 also foresees the possibility that if the cost involved in accommodating an individual or a
class of persons would bring about undue hardship for the employer, their exclusion or unequal
treatment may be considered to be based on a bona fide occupational requirement.

(d) Legitimate third-party preferences

Customers or persons with whom the employee must deal may have interests of privacy and
modesty that could justify exclusionary employment practices. For instance, it may be justifiable
that persons employed as custom officials who have to carry out strip searches be of the same sex
as those with whom they come into close physical contact when carrying out the duties for their
job.

In Sisley v Britannia Security Systems Ltd99, the employer’s claim was upheld that sex is a genuine
occupational qualification for a job that involved the workers being in a partial state of undress
while resting during long shifts. The same applies to a job as a sales assistant in a men’s clothing
shop involving physical contact in the form of taking inside leg measurements from male
customers.100

A problem which is sometimes encountered is the need to comply with foreign customs. Cases
decided in the US suggest that employers are not allowed to advance the need to comply with
foreign customs as a job requirement-related justification for discriminatory action.

In Fernandez v Wynn Oil Co101, the company contended that being male was an inherent job
requirement for a job that involved doing business with clients in South American countries. It was
alleged that the customers of clientele in South America would make it impossible for a woman to
succeed in such a position. The company argued, for example, that conducting business meetings
with a woman in her hotel room would offend South American attitudes. The court stressed that
stereotyped assumptions cannot form the basis of an inherent job requirement defence.

“Cultural necessity” is, however, accepted as a ground of justification for employment


discrimination in the European Community, Britain and Germany.102 This ground of justification

96 Op cit (1988) 289


97 International Union, UAW v Johnson Controls, Inc 111 SCt 1196 (1991) at 1209
98 Ibid

99 [1983] IRLR 404(EAT)

100 Wylie v Dee & Co (Menswear) Ltd [1978] IRLR 103

101 653 F2d 1273 (9th Cir 1981)

102 See e.g. s 7(2)(g) of the British Sex Discrimination Act, s 611a(1) of the German BGB.
45

applies where the employment involves work in a country in which it would be difficult for cultural
reasons for a woman to perform the work, because they are excluded from the public sphere and
from business activity in that country. The justification covers, for example, employment in some
Middle Eastern countries where Islamic law and the prevailing social mores render it more or less
impossible for women to carry out business activities.

Courts have consistently held that preferences of third parties, such as customers or co-workers,
for workers of a particular, gender, race religion, etc., which are not job related or are based on
unacceptable stereotypes regarding specific groups, will not satisfy the test for inherent job
requirements. One of the best known examples of the former instance is the case of Diaz v Pan
American Airways, Inc.103 The airline justified its policy of appointing only women as in-flight cabin
attendants because of the belief that they are better suited than men at “providing reassurance to
anxious passengers, giving courteous personalised service and, in general making flights as
pleasurable as possible within the limitations imposed by aircraft operations.” The court found in
favour of the excluded male complainant, because ministering to the psychological needs of the
passengers was tangential to the airline’s primary function of safely transporting passengers.

5.3 Operational justifications

5.3.1 Introduction

Unequal treatment in employment may be justified by legitimate concerns other than inherent job-
related requirements. Apart from job-related grounds of justification, prima facie employment
discrimination may be justified by reasons related to the general effective functioning of the
employer’s enterprise as a whole. For want of a better term, they will here be referred to as
operational justifications.104

Examples of operational justifications include: the maintenance of good industrial relations; the
observation of health and safety requirements105; the economic necessity of the workforce

103 653 F2d 1273 (9th CIR 1981).


104 The Labour Relations Act 1995 defines operational requirements in respect of dismissals, as the
economic, technological, structural, or similar needs of the employer. The Code of Good Practice on
Dismissal Bases on Operational Requirements.
105 New York City Transit Authority v Beazer 440 US 568 (1979). The safety of employees was also referred

to as an operational requirement in East Rand Proprietary Mines Ltd v UPUSA & Others [1997] 1BLLR
10(LAC) at 27
46

reduction or restructuring106; public credibility107; maintenance of a favourable public image108;


merit and productivity109; and the need to retain staff with scarce skills110.

The question whether operational considerations will be valid defences in particular circumstances,
will depend on whether they meet the test for legitimacy, fairness and justification as set out
earlier. Generally speaking, in order to be fair or justifiable, differential employment practices or
policies will have to serve legitimate and sufficiently compelling operational interests to be able to
justify its discriminatory effect. Furthermore, the normal proportionality considerations must be
applied, namely whether those practices are rationally related to achievement of the employer’s
operational ends and whether less discriminatory measures were available to achieve those ends.

In what follows next, some of the more important operational grounds of justification of differential
employment practices will be discussed.

5.3.2 Economic grounds of justification

Brodin describes the cost defence as a “justification offered by an employer other than that related
to the ability of persons to efficiently perform the particular job in question. Such justification
includes containment of costs in the areas of personnel administration, training, salaries and fringe
benefits, and avoidance of the risk of tort liability arising from workplace hazards.”111

As was pointed out in the previous section, accommodating previously excluded classes may
impose initial as well as ongoing direct or indirect costs on an employer. The court in British
Columbia (Public Service Relations Commission) v BCGSEU112 acknowledged as much when it is
stressed the fact that the duty of reasonable accommodation implies some hardship for the
employer, including economic cost short of expenses that would constitute undue hardship. It
follows that the mere fact that cost is involved in the implementation of non-discriminatory treatment
of employees, is not a valid reason not to implement equality in the workplace. In Dekker v
Strichting Vormingscetrum voor Jonge Volwassen (VJV Centrum)Plus113 the European Court of
Justice also held that the fact the employer would suffer financial detriment by not being reimbursed

106 Barnes v GenCorp Inc. 896 F2d 1457 (6th Cir 1990); Maresco v Evans Chemetics, division of WR Grace
& Co 964 F2d 106 (2d Cir 1992)
107 Davis v City of Dallas 777 F2d 205 (5th Cir 1985): as a justification for a requirement that applicants for

police officer positions should not have used marijuana recently or excessively.
108 Craft v Metromedia Inc 766 F2d 1205 (8th Cir 1985), concerning specific gender related dress codes for

newscasters.
109 In relations to wage differentials for instance

110 Bullock v Alice Ottley School [1992] IRLR 564 (CA); as justification for having different retirement ages

for different jobs, that disproportionally affect women employees


111 Brodin “Costs profits, and equal employment opportunity” 62 Notre Dame Law Review (1987) 318 at 320

n 14
112 (1999) 176 DLR (4th ) 1 at para 62

113 Case 177/8 [1990] ECR I-394l


47

by the state for the sickness benefits, which it would be obliged to pay the complainant during her
maternity leave, was insufficient to justify not appointing because she is pregnant.114

In Geller v Markham115 the court rejected the defendant’s cost-saving justification for a practice
indirectly discriminating against older workers. The defendant employer followed the practice of
recruiting teachers from people in lower pay grades with less than five years’ experience as
teachers. This resulted in a disproportional exclusion of older teachers whose longer experience
places them in a higher pay grade. The defendant sought to justify this policy as a necessary cost-
cutting gesture in the face of tight budgetary constraints. The court ruled that the cost justification
must fail. The direct or indirect exclusion of older workers was based on the general assumption
that the average cost of employing older workers as a group is higher than the average cost of
younger workers as a group, cannot be recognised as a valid justification for the discriminatory
policy. It would serve only to promote the very discrimination at which the Act was directed.116

The dismissal of a school bus driver over the age 65 because of the higher insurance cost for older
drivers was held unlawful in Tullis v Lear School.117 The court stated that increase insurance cost
is not a factor that would exempt the school from compliance with the duty not to discriminate
against older workers.118

Although the need to avoid increased cost will not ipso facto excuse unequal treatment, there are
obviously many instances where cost or other economic considerations could be relevant and
legitimate factors to take into account in assessing the fairness/ justifiability of an employer’s
conduct. Cost considerations are, for example, clearly relevant in respect of employee benefits.
The American Age Discrimination in Employment Act 1967 allows reasonable cost-based
differentiation between older and younger employees in respect of employer and/or employee
contributions to or coverage under certain benefit plans, such as life- and health insurance or
retirement schemes.119 The British Sex Discrimination Act 1975 allows cost as a relevant factor for
not appointing women in situations where workers have to live in premises provided by the
employer and it is not reasonable to expect the employer to equip its premises with separate
sleeping accommodation.

It is submitted that the factor of costs should only be allowed to justify otherwise discriminatory
practices in situations of real undue hardship for the employer.

The facts of McPherson v University of Montevallo120 is an example where a court found a


discriminatory practice justified by the necessity to save costs. The case concerned a claim of

114 Silke-Karin Mahburg v Land Mecklenburg- Vorpomern (ECJ) Case C-207-/98 (electronic copy:
<http://www.Europa.au.int/jurisp/egi-b…+T&séance+ARRET&where+())>: par 29.
115 635 F2d 1027 (2nd Cir 1980)

116 Id at 1034

117 874 F2d 1489 (11th Cir 1989) at 1490

118 29 CFR & 1625.7 (f)

119 29 USC & 623 (f) (2)

120 922 F2d 766 (11th Cir 1991)


48

indirect age discrimination because of the University’s policy of paying market-related salaries to
new appointees in the faculty of business, but not to incumbent professors. The court accepted as
a legitimate business reason for the salary differentiation the need to attract appropriately qualified
new staff in certain specialised fields in order to assure the faculty’s national accreditation. In
considering whether that purpose could have been achieved in a less discriminatory way, the court
took into account the factor of cost. It held that any alternative means of achieving the necessary
goal of accreditation must, at the very least be economically feasible for the employer. It found that
across-the-board market-based salary adjustments were not economically feasible at the time due
to the university’s precarious financial situation.121

Even in circumstances where economic considerations are compelling, the duty to consider less
discriminatory alternatives remains.

In SACCAWU v Garden Route Chalets122, the commissioner accepted that the extending of
transport benefits granted to some employees commuting from a neighbouring town, would impose
an unbearable financial burden on the employer. Although the exclusion of those employees from
the transport benefit has “a legitimate objective connected to business needs and economic factors
affecting the efficient carrying of its business”, the blanket refusal to extend the benefit is not a
necessary means of securing those ends.123 There are other compensatory options that may
amount to more proportional forms of differentiation. Instead of excluding a class of employees
from the benefit altogether, the employer could have granted them a travel allowance based on the
average monthly cost of the transport benefit received by other workers. Such an arrangement
would render the employee’s prima facie unfair conduct fair.124

5.3.3 Administrative burdens

121 922 F2d 766 (11th Cir 1991) at 733. See however, Benveniste v University of Southampton [1989] IRLR
122 (CA), where it was decided that if a change in the financial circumstances of the employer occurs at a
later stage, the economic justification for maintaining the pay differential may no longer apply. A university
lecturer was appointed at a time of severe financial constraints and agreed to a salary well below what she
would normally be entitled to. When the University’s financial situation improved and the constraints were
lifted, it failed to regrade her. The Court of Appeal rejected the university’s economic necessity defense,
based on the circumstances that obtained at the time of her appointment. It held that after the constraints
were lifted, there was no material difference between her and others doing work like work that could justify
the difference in pay. See also NUMSA v SAMANCOR Ltd (Meyerton works) (1998) 7 ARD 6.7.12. The
arbitrator questioned whether it is legitimate to sustain a wage differential indefinitely of whether the market
forces that compelled the initial differential still exist.
122 (1997) 1CCMA 4.2.1

123 Id at para 6-7

124 Id at para 7
49

Generally speaking, it seems that administrative difficulties associated with implementing equality
in the workplace would rarely be compelling enough to justify discriminatory employment practices.
The employer’s interest in avoiding administrative inconvenience does not represent a concern of
sufficient importance to tilt the scale in favour of maintaining a discriminatory practice or policy. 125
The duty of reasonable accommodation therefore also encompasses the bearing of administrative
burdens, short of undue hardship, associated with the implementation of equality in the workplace.

For instance, in Hardin v Stinchcomb126, the court rejected the idea that the administrative
inconvenience associated with the rearrangement of prison duties in order to accommodate
women guards in predominantly male prisons, justified their exclusion from such jobs.

In Merwine v Board of Trustees for the Institutions of Higher Learning127, the court noted that the
administrative inconvenience alone that would obtain in applying more subjective job requirements
could not justify imposing the formal educational requirements of a master’s degree for a job, which
disproportionately excluded women.

There are some examples where courts have considered administrative concerns important
enough to justify otherwise discriminatory employment practices.

In Rainey v Greater Glasgow Health Authority128, the House of Lords considered administrative
efficiency as a genuine need of the enterprise that justified differences in pay scales. The court
observed that legitimate grounds of justification for unequal treatment may include grounds “which
are other that economic, such as administrative efficiency in a concern not engaged in commerce
or business”.129 The complainant and other prosthetists, who entered the public health service
directly, were remunerated according to salary scales, which were lower than the rates paid to
those prosthetists (all male) who were initially recruited from the private sector. The latter were
paid more because of the necessity to attract privately employed prosthetists in order to establish
the new service. The court found that the new prosthetic service could not have been established
within a reasonable time if the earlier employees from the private sector had not been offered a
scale of remuneration no less favourable than that which they were enjoying as private
practitioners.130 That in view of the court, was undoubtedly a good objectively justified ground for
offering them the higher scale of remuneration. The lower scales of the prosthetists who were
recruited directly into the national health service were justified by the fact that “from the
administrative point of view it would have been highly anomalous and inconvenient if prosthetists
alone, over the whole tract of future time for which prosthetic services would endure, were to have
been subject to a different salary scale and a different negotiating machinery.” Accordingly, there
were “sound objectively justified administrative reasons” for placing prosthetists in general, men

125 See Blake v City of Los Angeles 595 F2d 1367 (9th Cir 1979) at 1376
126 691 F2d 1364 (11th Cir 1982)

127 745 F2d 631 (5th Cir 1985) at 640

128 [1987] 1AC 224 (HL)

129 Id at 38

130 Id at 239
50

and women alike, on the applicable National Health Service scale and subjecting them to its
negotiating machinery.131

5.3.4 Maintenance of sound labour relations / avoidance of labour unrest

The maintenance of sound labour relations and a stable workforce is an employer’s interest that
may in appropriate circumstances qualify as a legitimate ground of justification for otherwise
discriminatory employment practices.

The need to submit to established conventions and preferences in order to avoid industrial strife
may, however, not be allowed to perpetuate directly or indirectly discriminatory practices
embedded in fixed patterns of industrial regulation or preferences expressed by an employer’s
workforce. In the Australian Iron and Steel case, the consequence that the application of the
common “last in, first out” principle served to reinforce the discriminatory effect of past recruitment
practices, led the court to reject the fact that this was the established practice for redundancy
selection as a sufficient justification.132 Other instances mentioned by Hunter133 where courts have
invalidated the application of well-established and union-approved industrial principles because of
their discriminatory effect, include allocation of positions on the basis of seniority,134 a seniority-
based promotion system135and the practice of selecting part-timers first for redundancy.136

The fact therefore that a discriminatory practice is established through the normal process of
collective bargaining does not, as such, justify an employer implementing such a practice:
Leonard Dingler Employee Representative Council and others v Leonard Dingler (Pty) Ltd
and others.137

In the Danfoss Case138, the company paid the same basic salary to all employees in the same pay
grade. The collective agreement allowed pay differentials within specific grades based on the
criteria of flexibility, vocational training and seniority. The application of these criteria had the result
that the average pay of women was less than that of men within the same grade, a state of affairs
which the court found established a case of prima facie discrimination that required objective

131 Ibid. Cf the criticism of Hervey op cit 198


132 Australian Iron and Steel Pty Ltd v Banovic (1989) 168 (HC): par 27 of the judgment of Dawson J. He
added though that “the principle of ‘last in, first off” is not inherently unfair or unreasonable. All things being
equal, it may be a common sense way of selecting employees for retrenchment in the unfortunate event of
that being necessary…The requirement imposed in this case was unreasonable because in the particular
circumstances it repeated the discriminatory effect of the prior recruitment practice,”
133 Op cit 232

134 Steel v Union of Post Office Workers [1978] ICR 181 (EAT)

135 The proceedings Commissioner v Air New Zealand (1989) EOC 92-258

136 Clarke v Eley (IMI Kynoch Ltd [1983] ICR 165

137 (1997) 2LC 6.12.1 at 9-10

138 Handels-OG Kontorfunksionaerernes Forbund I Danmark v Dansk Arbejdsgiverforening (acting for

Danfoss) Case 109/88 (1989) IRLR 532 ECJ


51

justification from the employer. The fact that it was the result of a collective agreement did not in
itself justify the discriminatory practice.

5.3.5 Market forces

Employers often rely on market forces as a reason for unequal treatment of employees.139 One
important context in which market forces operate to create differential employment practices is
where a shortage of suitable candidates for a job forces employers to pay higher salaries to new
recruits, than to existing employees. In Clay Cross v Fletcher140, for example, a female sales clerk
was paid a lower salary than her male counterpart, who was appointed at a later stage. The
employer relied on the fact that he was the only suitable candidate at the time and that he would
have refused the job offer if he was not paid the same salary as at his previous job.

A shortage of candidates for a job was also considered sufficient justification for a pay differential
by the European Court of Justice in Enderby v Frenchay Health Authority and the Secretary of
State for Health.141 The complaint questioned the difference in pay between speech therapists
(predominantly women) and pharmacists and clinical psychologists (predominantly men) employed
by the National Health Service. The court held that the state of the employment market, which may
lead an employer to increase the pay of a particular job in order to attract candidates, might
constitute an objectively justified economic ground justifying the pay differential. The principle of
proportionality must be applied however, so that an increase in pay is justifiable only to the extent
that it is attributable to market forces. If it can be established what proportion of the increase in pay
is attributable to market forces, it may be accepted that the pay differential is objectively justified
to the extent of the proportion. If not, courts must assess whether the role of market forces in
determining the rate of pay was sufficiently significant to provide objective justifications for part or
all of the difference.

Some authors142 criticise these cases on the ground of its potential to limit the principle of equal
pay. The courts’ deference to market forces could, in opinion of the writers, have the effect of
maintaining traditional patterns of discrimination in pay between men and women or different races
performing work of equal value.

5.4 Public interest justification

Public interest-related justifications, according to Hervey, may arise in two contexts. The first, and
most common, concerns legislation or other state measures related to employment that results in
discrimination.143 These measures are justified if the purpose of the legislation, a general benefit to
the public or particular segment thereof, is judged to have greater importance than the equality

139 McCrudden. Equality in law between men and women in the European Community. United Kingdom
(1994) 50
140 NUMSA v Samancor [1998] 11 BALR 1501 (IMSSA)

141 Case 127/92 [1994] 1 All ER 495 (ECJ) at 524

142 Dine & Watt op cit 108 - 1099

143 Op Cit 201


52

rights of individuals or other groups.144 For instance, the United Kingdom government sought to
justify legislation excluding part-time workers from certain fringe benefits on the ground that
lowering the costs of part-time work would increase part-time job opportunities and thus all-over
employment levels.145

The second context is that in which a public interest justification is relied on by an employer, without
there being any statutory duty to adopt the discriminatory practice or policy in question. For
example, an employer may seek to justify an arrangement to provide special child care facilities or
benefits for employees without any caretaker-partner at home (which may benefit more female
than male employees), on the grounds that it serves a social need of its employees and their
children.

Public interest justifications will most commonly arise in the context of discriminatory legislation
applied in the workplace. Statutory authorisation for discriminatory employment measures is no
defence if challenged in terms of the constitutional equality provisions. The provisions of the
Employment Equity Act 1998 will also prevail over other legislation on any matter dealt with in the
Act, except an act of parliament expressly amending it.146 In terms of the constitutional framework
for equality, the fact that public interest justifications for legislation will concern broad issues of
public benefit may be an important relevant factor in judging its fairness or justifiability. It has been
shown that the factors for adjudging fairness or justifiability include the importance of the purpose
sought to be achieved by the discriminatory measure. Apart from this, there seems to be no reason
why public interest justifications are to be treated any differently from the other categories
discussed above. To adopt the same deference towards these forms of justifications as sometimes
expressed in other jurisdictions would therefore not fit into the present constitutional scheme. For
instance, in Nolte v Landeversicherunganstalt (Hanover)147the European Court of Justice observed
that a member state should be allowed a broad margin of discretion in choosing measures capable
of achieving the aims of their social employment policy. The standard of justification applied by
the European Court of Justice in R v Secretary of State for Employment, ex parte Seymour-Smith
& Perez148 seems more appropriate. The House of Lords referred to the European Court of Justice
the question regarding the legal criteria for establishing the objective justification of a measure
adopted by a member state in pursuance of its social policy. The United Kingdom government,
relying on the Nolte case, maintained that a member state should merely have to show that it was
reasonably entitled to consider that the measure would advance a social policy aim. The European
Court of Justice held, however, that although a member state has a margin of appreciation in
choosing the measures capable of achieving the aims of their social policy, this cannot have the
effect of frustrating the implementation of a fundamental principle of community law such as that
of equal pay for men and women. Only if a member state is able to show that the measures reflect
a necessary aim of social policy and are suitable and necessary for achieving that aim, will such

144 Id 105
145 R v Secretary of State for Employment, ex parte Equal Opportunities Commission and another [1995] 1
AC 1, [1994] ICR 317 (HL)
146 Section 63

147 Case 317/93 [1995] ECR I-4625 (ECJ) at para 33

148 Case C-167/97 [1999] 2 CMLR 273 (ECJ)


53

measures be held lawful in spite of their affecting more women than men in a negative way.
Relevant factors to be taken into account include whether the disputed measure is capable of
advancing that aim and the possibility of achieving that aim.149

There is very little judicial guidance on the more difficult question to what extent individual
employers, in the absence of any statutory authorisation, may rely on general public interest
justifications for discriminatory treatment. The problem is that such concerns may go far beyond
traditional employer defences related to inherent job and operational requirements. As appears
from the discussion of the inherent job requirements defence, courts generally insist that it should
be interpreted narrowly, in the sense of being necessary (or reasonably necessary) for the
performance of essential job functions. Operational requirements are contextualised within the
operational needs of the particular undertaking or enterprise. Neither inherent job requirements,
nor operational requirements are therefore normally associated with broader public interest or
social policy concerns. From a practical point of view, it would be clearly undesirable to allow
individual employers to justify exclusionary practices with reference to the kind of policy
considerations relied on in some of the cases mentioned earlier. For instance, the purposes of
employment discrimination legislation will be undermined if an individual employer, in the absence
of any statutory provisions on the matter, were to be allowed to justify the exclusion of certain
classes of workers from employment benefits on the basis that the reduced labour cost will open
up more job opportunities for that category of employees. On the other hand, the classic employer
defences need not be interpreted so narrowly that they unnecessarily inhibit social responsibility
concerns from justifying what otherwise might be a discriminatory measure, such as the example
of child care facilities mentioned earlier. Although such justifications may not be strictly speaking
related to inherent job or operational requirements in the traditional sense, they ought to be seen
to fall within the ambit of a socially broadened understanding of the employer’s duty of reasonable
accommodation (e.g. of family responsibilities). At least to this extent, it is submitted, should such
concerns be considered as legitimate grounds of justification.

6. Reasonable accommodation will be separately discussed in Unit 6.

7. Harassment will be separately discussed in Unit 5

149 Id at par 66-77 of the judgment.


54

UNIT 3

THE PROMOTION OF EQUALITY AND PREVENTION OF UNFAIR


DISCRIMINATION ACT

This part of the study guide does not represent original research. It consists of extracts taken or parts paraphrased from
the following sources: Albertyn et al Introduction to the Promotion of Equality and Prevention of Unfair Discrimination
Act, (2001); Benchbook for Equality Courts

1 BACKGROUND

The second important legislation issued in terms of section 9 of the Constitution is the Promotion of
Equality and Prevention of Unfair Discrimination Act, 4 of 2000. The main aim of the Act is to give
effect to the constitutional right to equality in all spheres of life, except in cases dealt with by the
Employment Equity Act. Like the Employment Equity Act, the Act prohibits unfair discrimination and
secondly requires the promotion of equality by way of active, positive measures that aim to advance
groups that have suffered discrimination and disadvantage. In addition, the Act prohibits hate
speech, harassment and the dissemination and publication of information that unfairly discriminates.
The latter aspects will be dealt with in other sections.

Regarding the wide scope of the Act, Albertyn et al (op cit 4) state the following:

"The Act allows for a very wide application that reflects a desire to reach deep into the private realm of
human interactions in preventing inequality. Many of these interactions have previously been regarded
as outside the scope of legal intervention. Thus, verbal sexist abuse, homophobic conduct or racial
harassment have previously been regarded as behavior which does not attract adequate legal sanction.
While the constitutional right to equality clearly has horizontal application, the Act makes this more explicit
and gives detail to the duty imposed on private actors by the right."

The wide reach of the Act is best illustrated in the schedule containing an illustrative list of unfair
practices in certain sectors. This list is attached at the end of this section.

2 STANDING

Section 20 provides:-

“Proceedings under this Act may be instituted by –

(a) any person acting in their own interest;

(b) any person acting on behalf of another person who cannot act in their own name;

(c) any person acting as a member of, or in the interest of, a group or class of persons;

(d) any person acting in the public interest;


55

(e) any association acting in the interest of its members;

(f) the South-African Human Rights Commission, or the Commission on Gender Equality.”

It appears from the above that the rules regarding standing closely follow the wording of section 38
of the Constitution. There are, however, notable differences between the Constitution and the Act
in this respect. Firstly, standing is explicitly afforded to the South African Human Rights Commission
and the Commission of Gender Equality to institute proceedings in equality cases. Secondly, the
Act has defines a " person” widely. “Person” is defined in section 1 and includes a juristic person, a
non-juristic entity, a group or category of persons. Therefore, the Act makes it possible not only for
natural and juristic persons to have standing in equality cases, but also for a non-juristic entity or a
group or category of persons. This means that some clubs, associations, committees, and so forth,
are afforded standing under the Act, even if they should not have legal personality.

The categories of persons with standing will now be discussed:

(a) Any person acting in their own interest

This is the traditional requirement for standing in South African law. Generally, courts have adopted
a restrictive attitude to the issue of standing, thus requiring that a person who approaches the court
for relief must have an interest in the sense of being personally affected by the wrong alleged.

b) Any person acting on behalf of another person who cannot act in their own name

A person or a group of people may be unable, because of their particular circumstances, to bring
action under the Act, for example, because of the physical location or due to intimidation, etc.

(c) Any person acting as a member, or in the interest of a group or class of persons

This makes provision for a so-called class action. As Chaskalson et al (Constitutional Law of South
Africa p. 8-7) states, the essence of a class action, or representative action as it is known in many
countries, is that one person may bring an action in the interest of a class of persons all having the
same cause of action. In other jurisdictions the representative plaintiff is being required to be a
member of the class in the sense that he or she shares the same cause of action and has the same
interest as the other members of the class. However, the use of the words "acting as a member of
or in the interest of " in section 20 of the Act makes it clear that the representative plaintiff/applicant
is not required to be pursuing his/her own interest.

(d) Any person acting in the public interest

This is the most far-reaching provision regarding standing under the Act, since its implication is that
an applicant will possess the necessary standing merely on the basis that he/or she is acting in the
public interest in seeking to enforce the Act. In Ferreira v Levin (1996 1 SA 984 CC) O’Regan J held
that the applicant has the onus to show that he/is genuinely acting in the public interest. She stated
further that:

"Factors relevant to determining whether a person is genuinely acting in a public interest will include
considerations such as: whether there is another reasonable and effective manner in which the challenge
56

can be brought; the nature of the relief sought, and the extent to which it is of general and prospective
application; and the range of persons or groups who may be directly or indirectly affected by any order
made by the Court and the opportunity that those persons or groups have had to present evidence and
argument to the Court."

An important difference between an action in the public interest and the abovementioned class
action, is that the former benefits members of a specific group, but is not binding upon them in terms
of the res judicata doctrine.

(e) Any association acting in the interests of their members

In the past, our courts had sometimes refused to allow associations to claim a relief on behalf of
their members, insisting that the individual members must approach the Court themselves. This
provision now explicitly allows all associations to institute proceedings on behalf of their members.

(f) Human Rights Commission, or the Commission on Gender Equality

This provision makes it possible for the SAHRC or the CGE to play a proactive role regarding the
prevention of discrimination. The SAHRC is an independent constitutional body with national
jurisdiction to investigate and to report on the observance of human rights and to take steps to
secure appropriate redress where human rights have been violated. Section 7 of the Human Rights
Commission Act further provides that the SAHRC may bring proceedings in competent courts or
tribunals in its own name, or on behalf of a person or a group or class of persons.

The CGE is an independent constitutional institution that promotes and advances gender equality.
It is a special institution with a broad range of powers and functions, including education,
investigation of complaints and monitoring of the state, the private sector and civil society on
progress made in respect of gender equality.

3 APPLICATION

Section 5 provides as follows:

1) This Act binds the state and all persons.

(2) If any conflict relating to a matter dealt with in this Act arises between this Act and the
provisions of the other law, other than the Constitution or an Act of Parliament expressly
amending this Act, the provisions of this Act must prevail.

(3) This Act does not apply to any person to whom and to the extent to which the Employment
Equity Act, 1998 (Act No. 35 of 1998), applies.

State:

The Act binds all organs of state, which include:

 any department of state or administration in the national, provincial or local sphere of


government;
57

 any other functionary or institution-

exercising a power or performing a function in terms of the Constitution or a provincial


constitution, or

exercising a public power of performing a public function in terms of any legislation or under
customary law or tradition.

This definition of state bodies is therefore considerably broader than the one contained in the
Constitution.The inclusion of functionaries and institutions exercising public power under customary
law or tradition is very significant, since this brings traditional bodies and officials into the fold of
state action. It is also notable that unlike the Constitution, the Act does not exclude courts and judicial
officers from the definition of "organ of state" These bodies are therefore by implication included in
the ambit of the Act.

Person:

The Act defines “person” to include a juristic person, non-juristic entity, a group or category of
persons. This wide definition has the consequence that the Act applies not only to natural and juristic
persons, but also groups and categories of persons without legal personality. Therefore,
partnerships and certain publicly important associations which may not comply with the
requirements of legal personality, are subject to the provisions of the Act.

4 JURISDICTION

The primary enforcement mechanism established by the Act are the equality courts. They are
specialized courts within the Magistrate and High Court with specific jurisdiction to hear complaints
under the Acts. (Albertyn op cit 18).

Generally, equality courts are subject to the same rules of jurisdiction as set out in the Magistrate's
Court Act 32 of 1944 and the Supreme Court Act 59 of 1959. A Magistrates Court sitting as an
equality court, however, is authorised by the Act to make a monetary order which exceeds the
jurisdiction of a magistrate’s court. Such an order must, however, be submitted to a judge of the
High Court for confirmation.

As it will appear more fully later, the jurisdiction of a Magistrate's Court sitting as an equality court is
also extended by the remedies which they are empowered to grant in terms of section 21(2) of the
Act. For instance, an equality court may issue a declaratory order even if it would have no jurisdiction
to do so when sitting as an ordinary magistrate's court. The same applies to the granting of
mandatory interdicts and orders for specific performance. An equality court is subject to the same
rules regarding the geographical jurisdiction of the ordinary Magistrate's or High Courts.

No court may function as an equality court and adjudicate equality claims unless and until a presiding
officer and a trained clerk are designated. The Act requires all presiding officers of equality courts
to undergo special training.

5 PROCEDURE
58

Guiding principles

The Act contains a number of guiding principles for procedure in equality courts. Amongst these are
included:

 accessibility and participation and access to justice,

 the expeditious and informal processing of cases (the Act emphasises that the enquiry must be
conducted in a speedy and informal manner. However, the equality courts are also courts of
record and the presiding officer must ensure that the proceedings are recorded at and certified),

 facilitation of participation of parties (The enquiry must also be conducted in a way that facilitates
participation, and where possible, in an environment that promotes participation. The rules of
procedure should be used and interpreted in a manner that promotes rather than inhibits
participation).

 access to justice to all persons in relevant forums (the principle of access to justice is particularly
important for unrepresented and/or indigent parties).

Laying a complaint

As was pointed out above, a wide variety of persons and groups have standing to bring complaints
under the Act. A person who wants to lay a complaint must notify the clerk of the equality court in
the prescribed manner. The details of this (e.g. standard forms, etc) are set out in the regulations.

Notifying the respondents

Notifying the respondent of the claim is important not only for informing him/her of the complaint, but
also to enable the respondent to provide information to the presiding officer to inform his/her decision
on the appropriate forum for dealing with the complaint. The issue of the form of service as well as
costs will be dealt with in the regulations.

Decision about the appropriate forum

Section 20(3)(a) provides that:

The Clerk of the Equality Court must, within the prescribed period of receiving such notification,
refer the matter to a presiding officer of the Equality Court in question, who must, within the
prescribed period decide whether the matter is to be heard in the Equality Court, or whether it
should be referred to another appropriate institution, body, court, tribunal or other forum
(hereafter referred to as an alternative forum) which, in the presiding officer’s opinion, can deal
more appropriately with the matter in terms of that alternative forum’s power and functions.

As appears from the above section, upon receipt of all relevant information from the complainant
and the respondent, the clerk of the equality court must refer the matter to the presiding officer, who
is then required to assess whether the matter can and should be dealt with by the equality court or
be referred to another forum. The alternative forums to which the presiding officers may refer the
59

claim include the following: The HRC, the CGE, the CCMA, the Land Claims Court and the Public
Protector.

The factors that the presiding officer must take into account when deciding whether to refer a claim
to an alternative forum are the following:

 The personal circumstances of the parties and particularly the complainant;

 The physical accessibility of any contemplated alternative forum;

 The needs and wishes of the parties and particularly the complainant;

 The nature of the intended proceedings and whether the outcome or the proceedings could
facilitate the development of judicial precedent and jurisprudence in this area of law.

 The views of the appropriate functionary at any contemplated alternative forum.

It is also required that the presiding officer must be of the opinion that the alternative forum can deal
with the matter more appropriately in terms of that alternative forum’s powers and functions. For
instance, where the dispute is capable of being settled by mediation, it would be appropriate for the
presiding officer to refer to the matter to such a mediating body (e.g. the CCMA, HRC or CGE).

Once the decision has been made to refer the matter to an alternative forum, it is the duty of the
clerk of the court to ensure that the referral takes place. The Act directs the alternative forum to
deal with the referred matter expeditiously in terms of its powers and functions.

The Act makes provision for an alternative forum to refer a matter back to the equality court. This
may take place where the alternative forum has failed to deal with the matter within a reasonable
period, or it is not able to resolve the matter to the satisfaction of one or both of the parties and one
or both parties so request.

Pleadings and pre-trial procedure

In so far as the Act or regulations do not provide for pleadings, procedures for obtaining documents
and other pretrial matters, the relevant rules of the Magistrate’s Courts and the High Court will apply.

Procedures during the hearing

As was stated above, the general principles regarding the procedures in equality courts require that
the hearing should be informal and participatory. The details of the procedures are spelled out in the
regulations, but it is clear that it is the intention of the Act that these procedures may deviate to a
substantial degree from the current practice. It is possible that the presiding officer will be required
to be more actively involved in establishing the facts of the enquiry and the liability of the parties.
This may assist in the creation of accessible, informal and participatory proceedings and might also
help to make the enquiry more fair to both parties, especially to disadvantaged parties who do not
have the resources to obtain legal representation (Albertyn 27).
60

6 REMEDIES

The Act requires the equality court or any other forum to provide effective remedies for the victims.
This will be the case if the remedy is adequate to deal with both individual as well as the systemic
forms of discrimination.

The range of remedies available to equality courts is the following:

(a) an interim order;

(b) a declaratory order,

(c) an order making a settlement between the parties to the proceedings an order of court,

(d) an order for the payment of damages in respect of any proven financial loss, including future
loss, or in respect of impairment of dignity, pain and suffering or emotional and psychological
suffering, as a result of the unfair discrimination, hate speech or harassment in question,

(e) after hearing the views of the parties, or, in the absence of the respondent, the views of the
claimant in the matter, an order for the payment of damages in the form of an award to an
appropriate body or organisation,

(f) an order restraining unfair or discriminatory practices or directing that specific steps be taken
to stop the unfair discrimination, hate speech or harassment;

(g) an order to make specific opportunities and privileges unfairly denied in the circumstances,
available to the complainant in question;

(h) an order for the implementation of special measures to address the unfair discrimination, hate
speech or harassment in question;

(i) an order directing the reasonable accommodation of a group or class of persons by the
respondent;

(j) an order that an unconditional apology be made;

(k) an order requiring the respondent to undergo an audit of specific policies or practices as
determined by the court;

(l) an appropriate order of a deterrent nature, including the recommendation to the appropriate
authority, to suspend or revoke the licence of a person;

(m) a directive requiring the respondent to make regular progress reports to the court or to the
relevant constitutional institution regarding the implementation of the court’s order;

(n) an order directing the clerk of the equality court to submit the matter to the Director of Public
Prosecutions having jurisdiction for the possible institution of criminal proceeding in terms of
the common law or the relevant legislation;
61

(o) an appropriate order of costs against any party to the proceedings;

(p) an order to comply with any provision of the Act.

Most of these remedies are self-evident and need not be discussed any further. There are some
innovative features about some of the remedies, however. Firstly, apart from a normal award of
damages, the Act also makes provision for compensatory damages that are punitive in nature.
Therefore, in the case of a particularly offensive form of discrimination, the court will be allowed to
order that a payment be made to an appropriate body or organization (e.g. bodies that are involved
in the combating of unfair discrimination). A punitive order of this kind is designed to deter future
violations of the right to equality, at the same time as it avoids unjustified financial benefits for the
complainant over and above compensation for the harm s/he has suffered.

A wide variety of different kinds of orders are available to address the specific problems associated
with discrimination, including an order to implement special measures, an order directing reasonable
accommodation of a particular group and an order requiring an audit of policies and practices of
offending individuals or institutions.

Albertyn (op cit 30) gives the following example of how the court may use a variety of possible orders
to address specific forms of discrimination.

A black woman brings a claim of unfair race discrimination to the equality court. She has been turned
down for occupancy of a townhouse in an upmarket area. During the course of the hearing it
emerges that she might not be the only black person who has been turned away by the respondent,
a large company of letting agents. The court finds that this is unfair race discrimination. What would
be an appropriate remedy? Here the court should not only consider a personal remedy for the
complainant (in the form of damages or an apology), but it could also order an audit of all the letting
practices of this particular estate agent, to address what seems to be a more systemic problem.
Additionally, the court could order the HRC to monitor compliance with this order by receiving regular
reports. The court could go still further, and alerted by what may be a discriminatory practice that
obtains in the letting sector as a whole, refer this aspect to the HRC for investigations.

7 APPEAL AND REVIEW

Section 23 of the Act provides that appeals may be made by any person aggrieved by any order
made by an equality court in terms of or under the Act. The appeal lies in the first instance to the
High Court having jurisdiction and thereafter to the Supreme Court of Appeal. In certain instances,
subject to the rules of the Constitutional Court, an appeal may be made directly to that court.

Manong & Associates (Pty) Ltd v Eastern Cape Department of Roads & Transport & others [2009]
JOL 23617 (SCA)

8 PROOF OF UNFAIR DISCRIMINATION


62

8.1 Introduction

The Act prescribes a different test for the determination of unfair discrimination than the Constitution.
As was explained earlier, the constitutional test for discrimination consists of three steps:

 does the impugned conduct amount to (prima facie) discrimination?

 if so, is the discrimination unfair?

 if so, is the unfair discrimination justified?

The Equality Act, on the other hand, envisages a two part enquiry:-

 does the impugned conduct amount to (prima facie) discrimination?

 if so, is the prima facie discrimination unfair?

Unlike the Constitution, therefore, the Act does not have a separate enquiry into justification. The
constitutional tests for unfairness and justification have been consolidated by the Act into one
enquiry regarding fairness.

The Act’s test for prima facie discrimination and unfairness will be discussed next.

8.2 Test for prima facie discrimination

The complainant firstly has to prove that the conduct of the respondent constitutes prima facie
discrimination. The elements of this prima facie case are found in the definition of discrimination in
section 1 of the Act.

Section 1 defines discrimination as follows:

Discrimination means any act or omission, including a policy, law, rule, practice, condition
or situation which directly or indirectly

(a) imposes burdens, obligations or disadvantage on; or

(b) withholds benefits, opportunities or advantages from,

any person or one or more of the prohibited grounds.

Discrimination, as defined in the Act, therefore, consists of three basic elements:

 an act or omission, which directly or indirectly,

 imposes a burden or withholds a benefit,

 on a prohibited ground.

“Act or omission, which directly or indirectly”


63

Discrimination can be a result of a positive act or an omission, i.e. a failure to act. Many cases of
discrimination involve a positive act of differential treatment that causes harm to a particular group.
Others may, however, involve discrimination which arises from a failure to differentiate in favour of
a particular group of persons. This is the form that indirect discrimination usually takes. It is accepted
that a failure to differentiate between groups whose circumstances are materially different is a major
contemporary form of unfair discrimination. An example of discrimination by omission can be found
in the failure of state hospitals to provide sign language interpretation for deaf patients.
Differentiation in favour of the group of disabled people is required in order to recognize and address
the material differences that they experience in accessing the health system and the likely unfair
impact of the perpetuation of these differences.

As appears from the definition of discrimination, an act or omission may include a policy, law, rule,
practice, condition on situation. This all-inclusive definition ensures that the meaning of
discrimination goes beyond individual acts to include the broader social and structural causes of
discrimination. As Albertyn (op cit 33) explains, "practice, condition and situation" may refer to the
unwritten and often hidden rules, forms of behavior and attitudes that may cause or exacerbate
discrimination. The expanded definition seeks to indicate that the hidden barriers that cause unfair
discrimination, as well as the failure to address an ongoing discriminatory situation, can amount to
unfair discrimination.

“imposes a burden or withholds a benefit”

Discrimination always involves some element of harm or prejudice. In terms of the Act, this can arise
either from the imposition of burdens, obligations or disadvantage or the withholding of benefits,
opportunities or advantages. The complainant bears the onus to establish harm in any of the two
above forms. An example of the first form can be found in Harksen v Lane (1997 11 BCLR 1489
CC), where the burden on a solvent spouse of having her property vest, together with that of her
insolvent spouse, with the trustee of the estate of the insolvent spouse, was found to constitute
prima facie discrimination. An example of discrimination in the form of the denial of benefits or
advantages, is in the case of President of the RSA V Hugo (1997 6 BCLR 708 CC) where a father
was excluded from the application of a pardon that was granted to prisoners who were mothers.

“on a prohibited ground”

The discrimination must occur on a prohibited ground or on a combination of those grounds.


Prohibited grounds may be those listed in the Act or unlisted ground that that are analogous to the
listed grounds.

The Act lists the usual number of prohibited grounds: race, gender, sex, pregnancy, marital status,
ethnic or social origin, colour, sexual orientation, age, disability, religion, conscience, belief, culture,
language and birth.

If the complainant seeks to prove that the discrimination took place on an unlisted ground, he or she
must show that this unlisted ground meets criteria set out in Section 1 (xxii) (b) of the Act. In terms
of these criteria, it must be shown that discrimination based on that ground:

 causes or perpetuates disadvantage;


64

 undermines human dignity; or

 adversely affects the equal enjoyment of a person’s rights and freedoms in a serious manner
that is comparable to discrimination on a listed ground.

It appears that this test is similar, but not quite the same as the test for unlisted prohibited grounds
of discrimination in terms of section 9 of the Constitution. While the Constitutional Court has thus far
used the potential to impair dignity as its main criterion, the Act provides three alternative criteria.
The complainant needs only to prove that one of these criteria applies.

8.3 Burden of proof regarding the determination of discrimination

Section 13 (1) of the Act provides as follows:

If the complainant makes out a prima facie case of discrimination:

(a) the respondent must prove, on the facts before the court, that the discrimination did not take
place as alleged; or

(b) the respondent must prove that the conduct is not based on one or more of the prohibited
grounds.

This section requires the complainant to make out a prima facie case of discrimination. It also sets
out the manner in which the respondent may escape liability and prove that there was no
discrimination.

Firstly, the respondent can prove that on the facts before the court, discrimination did not take place
as alleged. In other words, he/or she may disprove theact or omission claimed.

Secondly, the respondent may avoid liability if he or she is able to prove that the conduct was not
based on one or more of the prohibited grounds. In alleging a link between the discrimination
complained of and a prohibited ground, intention is not required to show either direct or indirect
discrimination. It is also sufficient if the discrimination is partly based on the prohibited ground. A
claim should not be rejected at this stage because there were other reasons for discrimination,
because the ground was not the sole reason for the discrimination, or because the discrimination
had a non-discriminatory purpose. These issues only become relevant to the enquiry into fairness,
where the court enquires into the nature, extent and purpose of the discrimination.

8.4 Proof of unfairness

Section 13(2) of the Act provides as follows:

If the discrimination did take place-

a) on a ground in paragraph (a) of the definition of "prohibited grounds", then it is unfair,


unless the respondent proves that the discrimination is fair;

b) on a ground in paragraph (b) of the definition of "prohibited grounds", then it is unfair-


65

i) if one or more of the conditions set out in paragraph (b) of the definition of "prohibited
grounds" is established; and

ii) unless the respondent proves that the discrimination is fair.

Section 14 of the Act provides as follows:

(1) It is not unfair discrimination to take measures designed to protect or advance persons or
categories of persons disadvantaged by unfair discrimination or the members of such groups or
categories of persons.
(2) In determining whether the respondent has proved that the discrimination is fair, the
following must be taken into account:
(a) the context;
(b) the factors referred to in subsection 3;
(c) whether the discrimination reasonably and justifiably differentiates between
persons according to objectively determinable criteria, intrinsic to the activity concerned.
(3) The factors referred to in subsection (2)(b) include the following:
(a) whether discrimination impairs or is likely to impair human dignity;
(b) the impact or likely impact of the discrimination on the complainant;
(c) the position of the complainant in society and whether he or she suffers from
patterns of disadvantage or belongs to a group that suffers from such patterns of disadvantage;
(d) the nature and extent of the discrimination;
(e) whether the discrimination is systemic in nature;
(f) whether the discrimination has a legitimate purpose;
(g) whether and to what extent the discrimination achieves its purpose;
(h) whether there are less restrictive and less disadvantageous means to the achieve
the purpose;
(i) whether and to what extent the respondent has taken such steps as being
reasonable in the circumstances to:
(i) address the disadvantage which arises from or is related to one or more of the
prohibited grounds; or
(ii) accommodate diversity.

We will discuss each of the above in sequence.

8.4.1 Positive measures for protection and advancement

Section 14 (1) states that;

"It is not unfair discrimination to take measures designed to protect or advance persons or categories
of persons disadvantage by unfair discrimination or the members of such groups or categories of
persons."

This is a repetition of the provision in section 9(2) of the Constitution. Affirmative action will be
separately discussed and will therefore not be discussed here.
66

8.4.2 The enquiry into unfairness in terms of section 14(2) and (3)

Sections 14(2) and (3) list the criteria that must be taken into account in determining whether the
discrimination is fair, namely:

 context,

 the factors listed in section 14 (3),

 the requirements of section 14 (2) (c).

(a) Context

The Act requires that the "context" must be taken into account. Context here refers to a method of
interpretation that seeks to understand the overall impact of discriminatory action in the context of
people’s lives. This includes an examination of the history and socio-economic conditions of the
individuals and groups concerned, as well as the impact of the impugned provision on social patterns
and systemic forms of disadvantage. The equality courts are, therefore, required to consider the
impact of the alleged discrimination on the litigant, as well as on his/her group. Albertyn (op cit 42)
states that a broad range of evidence will be relevant to this enquiry, including statistical and
sociological evidence as well as socio-economic analysis. Judicial officers will be required to ensure
that they are able to step outside their own experiences and critically consider situations that they
may not have previously encountered.

(b) Factors listed in section 14(3)

The Act requires Courts to consider a list of nine factors set out in section 14(3) in the inquiry into
unfairness. According to Albertyn (op cit 43) these factors can be divided into two broad categories,
namely-

 factors that involve an assessment of the impact of the discrimination on the complainant
and his/her group on the one hand, and

 factors that focus on the behaviour and justifications of the respondent (which involves an
assessment of the purpose of the discrimination, whether the purpose is legitimate and the
relationship between the discrimination and its purpose).

We will briefly look at each of them in turn.

(i) Whether the discrimination impairs or is likely to impair human dignity

In the first section, it was stated that the constitutional Court consider discrimination to be essentially
a form of impairment of a person’s dignity.
67

The Supreme Court of Canada in the case of Law v Canada (1999) 170 DLR 4th 1 (SCC) describes
human dignity as follows:-

"Human dignity means that an individual or group feels self-respect and self-worth. It is concerned with
physical and psychological integrity and empowerment. Human dignity is harmed by unfair treatment
premised upon personal traits or circumstances which do not relate to individual needs, capacities or
merits. It is enhanced by laws which are sensitive to the needs, capacities and merits of different
individuals, taking into account the context of their differences. Human dignity is harmed when individuals
and groups are marginalised, ignored, or devalued, and enhanced when laws recognise the full place of
all individuals and groups within the Canadian Society."

(ii) The impact or likely impact of the discrimination on the complainant

The constitutional test for unfairness emphasises that the impact of discrimination on the
complainant is the cardinal determining factor. This factor requires a court to examine the actual
consequences of discriminatory conduct for the complainant and his/her group.

(iii) The position of the complainant in society and whether he or she suffers from patterns
of disadvantage or belongs to a group that suffers from such patterns of disadvantage

This is also one of the factors to be considered in the determination of fairness in terms of the
constitutional test for unfair discrimination. It has already been discussed above.

(iv)The nature and extent of the discrimination

This factor requires the Court to consider issues such as whether the discrimination is limited to the
complainant or whether it affects a broader group, whether it was a single incident or an ongoing
situation, as well as the extent to which it affects the complainant and his or her rights or interests.
The court could also consider whether the discrimination was a necessary part of a broader scheme
to overcome discrimination.

(v) Whether the discrimination is systemic in nature

Systemic discrimination is particularly deep-rooted in society and often has multiple consequences
for the person who suffers discrimination. Thus women's role in caring for children has had various
negative effects in the public sphere (workplace, politics, and economy). Such forms of
discrimination may be so imbedded in every-day life that it has almost become invisible.

(vi) Whether the discrimination has a legitimate purpose

As was pointed out in the discussion of the constitutional test for unfair discrimination, in order to be
justifiable, the discriminatory conduct must have a worthy and important and purpose. For instance,
in the earlier discussion of justifications for employment discrimination, the issue was raised whether
commercial interests as such can be important enough to outweigh the disadvantageous effect of
discriminatory conduct. The European Court of Justice refused to accept that an increase in the
68

costs of labour is important enough to justify the employer’s refusal to appoint a pregnant woman. 150
The same was said in SAA v Hoffman in respect of the airway’s refusal to appoint HIV positive
persons as cabin attendants. SAA argued that it might suffer a commercial disadvantage vis a vis
its competitors should it be forced to change its policy. The court noted that –

“We must guard against allowing stereotyping and prejudice to creep in under the guise of commercial
interest. The greater interests of society require the recognition of the inherent dignity of every human
being, and the elimination of all forms of discrimination. Our Constitution protects the weak, the
marginalised, the socially outcast, and the victims of prejudice and stereotyping. It is only when these
groups are protected that we can be sure that our own rights are protected.”

(vii) Whether and to what extent the discrimination achieves its purpose

This factor has also been discussed above in the section dealing with the constitutional test for
discrimination.

(viii) Whether there are less restrictive and less disadvantageous means to achieve the
purpose.

See the discussion above in the section dealing with the constitutional test for unfair discrimination.

(ix) Whether and to what extent the respondent has taken such steps as being reasonable in
the circumstances to-

(i) address the disadvantage which arises from or is related to one or more of the
prohibited grounds; or

(ii) accommodate diversity.

This factor requires the Court to consider whether the respondent has taken reasonable steps to
address the disadvantage caused by the discrimination or to accommodate diversity. This means
that where a court finds that discriminatory conduct has occurred, but where the respondent can
show that it has initiated a process aimed at addressing inequality and discrimination, the alleged
discrimination may be fair

(c) The requirement of section 14(2)(c)

This section provides that in determining whether the respondent has proved that the discrimination
is fair, it must be taken into account-

Whether the discrimination reasonably and justifiably differentiates between persons according
to objectively determinable criteria, intrinsic to the activity concerned.

150
Dekker v Stichting Vormingscentrum voor Jonge Volwassen(VJV Centrum) Plus Saakno 177/88 (1990)
ECR I-3941.
69

This factor is explained by Albertyn (op cit 46) with reference to the example of a black African
waiter, with 20 years experience, being turned down in favour of a waiter of Chinese origin (with 2
years experience), for a job at a Chinese restaurant. The respondent in this case would have to
show that there is a causal link between the objectively determinable criterion (race or ethnicity in
this example) and the relevant activity (in this case the business of running a Chinese restaurant).
The respondent would have to show that the ethnicity of the waiters is essential to the authentic
atmosphere and successful business of running a particular kind of restaurant. The use of the word
“intrinsic” suggests that the relationship between the criterion of differentiation (i.e.race/ethnicity)
and the activity (running a Chinese restaurant) should be so close that the activity would not be
possible if the criterion were not adopted. The section also requires the respondent to show that
such a differentiation is reasonable and justifiable in the circumstances, i.e. that the respondent’s
need to make the differentiation outweighs its negative effect on the complainant

APPLYING THE ACT’S TEST FOR UNFAIRNESS

The following examples are taken from Albertyn (op cit 48):

Example 1

Pumla’s husband dies. Her brother-in-law is awarded the family home in line with the regulations
governing the customary law of succession. She brings a claim of unfair discrimination on the basis
of gender.

Section 14(2) context: Requires that the matter should be adjudicated by understanding the actual
impact of this rule not only on Pumla and her family, but on women generally. It should not be
determined abstractly by referring to the law as it appears in the books, but to how it operates in
practice. This, in turn, requires information on the actual practices of inheritance, and certainly
requires sociological or socio-legal information on these practices.

Section 14(2) and (3)(a)-(e): These sections require the court to look at what happens to Pumla and
women like her in such a situation. It will hear evidence that despite a customary obligation of
maintenance, the reality is that Pumla will probably be thrown out of her home with no redress. This
causes enormous disadvantage to her and many women who have experienced this before het.
She loses her home and often her belongings. Often she has contributed financially to this home
and thus she suffers material disadvantage. Moreover her dignity is impaired by the fact that she is
not deemed fit to inherit as she is not a man.

The discrimination is both widespread and systemic, based on deeply rooted and patriarchal notions
of women’s position in the family. Women are seen to be incapable of managing property.

Section 14(2) and (3)(f)-(i): Here the court will have to look at whether there is a legitimate purpose
to the law. This may be that immovable property belongs to the family and not to individuals and that
men are heads of families and women are looked after within customary law settings. The court will
have to weigh this against the reality that women are often thrown out of their homes and are not
maintained by the male heir. If the court has to consider less disadvantageous means of meeting
70

the collective purpose of providing for needy members of the family, it can decide that women are
equally able to address the needs of dependent family members (and more likely to given their role
in respect of children).

Section 14(2)(c): While sex/gender may be an objectively determinable criterion, the exclusion of
women from inheritance cannot be seen to be intrinsic to any identified activity. Although this section
must be considered, it does not seem relevant to an enquiry of this nature and will not adduce any
new information.

Example 2

Frank claims that an insurance company has unfairly discriminated against him when it refused him
medical insurance. Frank is 65 and a smoker.

Section 14(2)(a) context: This requires an analysis of the provision in the context of the situation of
old people, their ability to obtain insurance and the impact of such restrictions on the group.

Section 14(2)(b) and (3)(a)-(e): Here the court looks at the impact of the refusal of insurance on the
basis of age and health status. It may look at how this refusal denies people who are able to pay
for private insurance access to that type of cover. It may conclude that this forces more people into
the overcrowded public health sector.

Section 14(2)(b) and (3)(f)-(i): The court would listen to the explanations of the insurance company
in relation to risk and the costs associated with levels of risk and effective insurance for the majority
of people. It would have to scrutinise this stated purpose and balance it against the extent of the
discrimination and alternative ways (if any) of assessing risk. It would also consider the extent to
which there were other forms of insurance available to Frank from the same company.

Section 14(2)(c): This would involve a similar process to that in s 14(2). The court would look for an
objectively identifiable criterion (age and health status) and determine whether it was intrinsic to the
activity (risk in insurance). Having found this, it would have to decide whether the differentiation on
the basis of age and health status was reasonable and justifiable. Seeking guidance from the Act
and the Constitution, it would find that it has to engage in a balancing process with the criteria of s
14(2)(b). In essence it would carry out the same exercise twice. To simplify the process – it would
deal with 14(2)(b) and (c) as one process.

Schedule

ILLUSTRATIVE LIST OF UNFAIR PRACTICES IN CERTAIN SECTORS (Section 29)


1. Labour and employment
(a) Creating artificial barriers to equal access to employment opportunities by using certain
recruitment and selection procedures.
(b) Applying human resource utilisation, development, promotion and retention practices
which unfairly discriminate against persons from groups identified by the prohibited grounds.
(c) Failing to respect the principle of equal pay for equal work.
71

(d) Perpetuating disproportionate income differentials deriving from past unfair discrimination.

2. Education
(a) Unfairly excluding learners from educational institutions, including learners with special
needs.
(b) Unfairly withholding scholarships, bursaries, or any other form of assistance from learners
of particular groups identified by the prohibited grounds.
(c) The failure to reasonably and practicably accommodate diversity in education.

3. Health care services and benefits


(a) Subjecting persons to medical experiments without their informed consent.
(b) Unfairly denying or refusing any person access to health care facilities or failing to make
health care facilities accessible to any person.
(c) Refusing to provide emergency medical treatment to persons of particular groups
identified by one or more of the prohibited grounds.
(d) Refusing to provide reasonable health services to the elderly.

4. Housing, accommodation, land and property


(a) Arbitrary eviction of persons on one or more of the prohibited grounds.
(b) “Red-lining” on the grounds of race and social status.
(c) Unfair discrimination in the provision of housing bonds, loans or financial assistance on
the basis of race, gender or other prohibited grounds.
(d) Failing to reasonably accommodate the special needs of the elderly.

5. Insurance services
(a) Unfairly refusing on one or more of the prohibited grounds to provide or to make available
an insurance policy to any person.
(b) Unfair discrimination in the provision of benefits, facilities and services related to
insurance.
(c) Unfairly disadvantaging a person or persons, including unfairly and unreasonably refusing
to grant services, to persons solely on the basis of HIV/AIDS status.

6. Pensions
(a) Unfairly excluding any person from membership of a retirement fund or from receiving any
benefits from the fund on one or more of the prohibited grounds.
(b) Unfairly discriminating against members or beneficiaries of a retirement fund.

7. Partnerships
(a) Determining in an unfair discriminatory manner who should be invited to become a
partner in the partnership in question.
(b) Imposing unfair and discriminatory terms or conditions under which a person is invited or
admitted to become a partner.

8. Professions and bodies


72

(a) Imposing conditions that unfairly limit or deny entry into the profession of persons from
historically disadvantaged groups.
(b) Unfairly limiting or denying members access to benefits or facilities on the basis of a
prohibited ground.

9. Provision of goods, services and facilities


(a) Unfairly refusing or failing to provide the goods or services or to make the facilities
available to any person or group of persons on one or more of the prohibited grounds.
(b) Imposing terms, conditions or practices that perpetuate the consequences of past unfair
discrimination or exclusion regarding access to financial resources.
(c) Unfairly limiting access to contractual opportunities for supplying goods and services.

10. Clubs, sport and associations


(a) Unfairly refusing to consider a person’s application for membership of the association or
club on any of the prohibited grounds.
(b) Unfairly denying a member access to or limiting a member’s access to any benefit
provided by the association or club.
(c) Failure to promote diversity in selection of representative teams.
73

UNIT 4

HARASSMENT

Study the following pages from Albertyn et al 2001: 98-99.

1. Introduction

Section 6(3) of the Employment Equity Act provides:

Harassment of an employee is a form of unfair discrimination and is prohibited on any one, or a combination
of the listed grounds.151

Employers have a duty to prevent and react to instances of harassment. This entails
implementing harassment policies with sufficient preventative measures, as well as instituting
effective procedures and mechanisms for dealing with harassment in the workplace.

South African law is well developed in respect of sexual harassment but not in respect of other
types of harassment. However, harassment based on the other prohibited grounds of
discrimination is apt to be governed by the same legal principles governing sexual harassment.
For this reason, this section will focus mainly on sexual harassment.152

Harassment is also prohibited by the Promotion of Equality and Prevention of Unfair Discrimination
Act 2000. Nedlac has also published a Sexual Harassment Code in terms of section 203(1) of the
Labour Relations Act 1995. The provisions of these two instruments must therefore be taken into
account together with the EEA.

2. Sexual harassment

2.1 What is sexual harassment?

151 These include race, gender, sex, pregnancy, marital status, family responsibility, ethnic or social origin,
colour, sexual orientation, age, disability, religion, HIV status, conscience, belief, political opinion, culture,
language and birth. Since discrimination on non-listed grounds is also possible, following the discrimination
test set by the Constitutional Court, it is submitted that harassment could be committed on such grounds as
well. It could, for example, include citizenship and physical characteristics.
152 Harassment based on sexual orientation is another prevalent form of workplace discrimination. See

Higgins v New Balance Athletic Shoe Inc 21 F Supp 2d 66 (D Me 1998).


74

Sexual harassment is defined in the Nedlac Sexual Harassment Code (See below) as:

“unwanted conduct of a sexual nature”153

The Australian Sexual Harassment Code154 defines sexual harassment as:

“an unwelcome sexual advance, unwelcome request for sexual favours or other unwelcome
conduct of a sexual nature which makes a person feel offended, humiliated or intimidated,
and where that reaction is reasonable in the circumstances.”

In the case of J v M Limited155, which has been followed in numerous sexual harassment cases,
sexual harassment was defined very broadly as “unwanted sexual attention in the employment
environment … which has a negative effect on the recipient.”

Thus, welcome and mutual sexual behaviour is excluded from the definition.

When does sexual attention become harassment? US courts have emphasized that sex
discrimination law does not prohibit an attempt to establish personal relationships, even if this
occurs in the workplace. So, a cause of action does not normally arise from an isolated incident
or a mere flirtation (Heelan v Johns-Manvilee Corp. 451 F Supp 1382). The difficulty is to draw the
line between harassment and permissible sexual advances or attention. Item 3(2) of the Sexual
Harassment Code states that sexual attention becomes harassment if –

 the conduct is persisted in, and/or

 the recipient clearly indicated the offensive nature of the behaviour, and/or

 the perpetrator should have known that the behaviour is regarded as unacceptable.

The Promotion of Equality and Prevention of Unfair Discrimination Act requires:

 persistency or seriousness and

 that the behaviour demeans the victim or creates a hostile environment.

Because sexual harassment is essentially unwanted conduct of a sexual nature, it is not limited
to persons of the opposite sex. It was thought previously that sexual harassment had to take
place within the “ordinary” context of sexual interest between men and women. This notion has
now effectively been dispelled by case law. In the United States, the first class action of male-on-
male sexual harassment has been settled by the Equal Employment Opportunity Commission,
after the 1998 Supreme Court case of Oncale v Sundowner Offshore Services Inc.156 The

153 Item 3.
154 Human Rights and Equal Opportunities Commission (Australia) op cit (1996) 4.
155 (1989) 10 ILJ 755 (IC) at 757 D-G.

156 US 118 SCt 998, 1003 (1998).


75

harassment of a female assistant by a female supervisor was found to constitute sex discrimination
in the United Kingdom.157

Sexual harassment must be distinguished from sex-based harassment. Unlike sexual


harassment, sex-based harassment does not necessarily include the aspect of sexuality.158 Sex-
based harassment can be defined as:

“adverse or demeaning language or conduct generally indicating hostility in the presence


of persons in the workplace because of their sex.”159

It includes derogatory remarks such as “women are hysterical”, “this is a man’s job”, “only a stupid
female” and “bitch”. A remark about a woman’s breasts constitutes sexual harassment, whilst one
on a man’s balding head doesn’t.160 The last-mentioned remark may, however, constitute sex-
based harassment. It’s not the sex of the victim itself that is the determining factor, but rather the
sexual nature of the conduct and (perceived) sexuality of the victim.

It is also possible that the harassment focuses more on gender, i.e. assumptions as to the social
role or behaviour of women. The manner in which a complainant of sex discrimination in Price
Waterhouse v Hopkins161 was treated when she was denied a partnership could serve as an
illustration. It was suggested to her that she try charm school, a different dress style and improved
manners for “a more feminine approach”.

Canadian law also provided for “personal harassment”, i.e. unwelcome behaviour not based on
any of the prohibited grounds, but on the abuse of authority aimed at the specific person.162

2.2 Forms of sexual harassment

Usually, two basic types of sexual harassment are distinguished, namely:

 quid pro quo harassment, and

 hostile work environment harassment.

(a) Quid pro quo harassment

Quid pro quo harassment occurs when a person is required to engage in or tolerate sexual
acts in order to obtain job-related benefits or avoid job-related detriment.

157 Johnson v Gateway Food Market Ltd COIT 4079/90, discussed in Hemming op cit 262.
158 Equal Employment Opportunities Commission (US) Notice N-915-150 of 3/19/90
<www.eeoc.gov/docs/currentissues.html> 9.
159 EEOC and Ford signs multi-million settlement of sexual harassment case (http://www.eeoc.gov/press/9-

7-99.html) par 10.


160 Institute Cleaning Co Ltd v Heads [1995] IRLR 4 EAT.

161 (1989) 109 SCt 1775.

162 Canadian Human Rights Commission op cit (1998).


76

The essence of quid pro quo harassment is a situation where an individual must choose between
foregoing job benefits or submitting to sexual demands.

As it is often the agents of employers (managers, supervisors, etc.) that in the ordinary course of
work award employment advantages, vicarious liability is likely to ensue in cases of quid pro quo
harassment in South Africa. Employers, therefore, have to ensure that those making employment
decisions in relation to any employment practice listed in s 1 Employment Equity Act are aware of
the prohibition on harassment and the consequences where in such employment practices benefits
are made subject to submission to sexual harassment.

According to Tiger Wheels Manufacturing and NUMSA v Dikgang163, it does not matter that the
harasser did not actually have the power to dismiss, refuse to employ or transfer employees who
refused his sexual advances. The fact that the women believed that he indeed had the power to
influence the course of their employment if they did not agree to his sexual advances, was
considered an aggravating factor.

(b) Hostile work environment

Sexual harassment can create an unpleasant or even unbearable work environment for the victim,
even if it is not related to any job benefit.

In this form of harassment, the victim’s employers, supervisor, co-workers, or even non-employees
(such as clients) engage in sexually related conduct which the victim finds offensive or demeaning.
Such conduct usually involves a series of incidents or a pattern of conduct rather than a single
episode with one or more of the following consequences-

 the victim’s work performance is detrimentally affected, or

 an intimidating, hostile or offensive working environment

The essence of hostile environment sexual harassment is that the victim must endure a work
environment that, while not necessarily causing any economic loss, causes psychological or
emotional harm or otherwise unreasonably interferes with the victim’s job performance.

Forms of hostile environment sexual harassment

Sexual advances. A common example is where female employees are subjected to


unwelcome sexual advances, innuendo, touching and propositions by male employees
who may or may not be supervisors. The conduct, which would not necessarily be
considered unwelcome by all women, continue, although the victim makes it clear that it
is unwelcome to her. The women, who may or may not eventually submit, suffer
emotional distress and find it difficult to do their job.

Gender-based animosity. In another common situation, a woman employee, often in a


traditionally male-dominated occupation (e.g. truck driving) or workplace (e.g. a gold

163 (1996) 5 ARB 8.25.10: <http://www.irnet.co.za.>


77

mine), is subjected to scorn, ridicule, and verbal abuse from males who resent her
presence. The behaviour consists of gestures, words, or conduct that is sexually related.

Sexually charged workplace. In a third situation, a woman’s workplace features open


displays of phonographic posters, calendars, and graffiti, or pervasive incidents of sexual
horseplay or sexual favouritism. This conduct, even if not directed specifically towards
the woman, may create a hostile work environment for her, or for women employees in
general.

Sexual favouritism. Item 4(2) of the Sexual Harassment Code defines “sexual
favouritism” as the situation in which a person in a position of authority rewards only those
responding to his/her sexual advances. In the typical sexual favouritism scenario,
supervisors favour subordinates who grant sexual favours, thereby creating an implicit
quid pro quo. The victim of sexual favouritism is not only the one who has been subjected
to sexual advances in order to be granted employment benefits, but also the person (a
third party) whose job opportunities are negatively affected by sexual favouritism of an
employer/manager towards another employee, without the third party having herself been
sexually harassed. If the sexual favouritism is pervasive, it constitutes hostile
environment sexual harassment.

2.3 Elements of sexual harassment

In South African cases164 the following have been set as “requirements” for conduct to constitute
sexual harassment:

 unwanted conduct,

 of a sexual nature (that may be verbal, non-verbal or physical),

 which is persistent or serious,

 affects the dignity of the harassed person.

The following discussion will deal with the elements most pertinent in current legislative definitions
and case law.

(a) “Unwanted conduct”

164See Pick ‘n Pay Stores Limited v an Individual (1994) 3 (1) ARB 8.25.136; Transtel Johannesburg and
TWU (1998) 7 ARB 8.25.3: <http:/www.irnet.co.za>. It was, however, stated in the Pick ‘n Pay case that the
“crucial element” is proving the conduct as “unwanted”. This is in line with the approach in the Sexual
Harassment Code. In many other sexual harassment cases the conduct at issue was found to have
constituted sexual harassment without it being tested against the five aspects.
78

Both the Promotion of Equality and Prevention of Unfair Discrimination Act and the Sexual
Harassment Code refer to “unwanted” conduct. The Sexual Harassment Code explicitly
distinguishes harassment from conduct that is welcome. The words “unwanted” and “unwelcome”
have the same meaning and can be used interchangeably.

In the United States, “unwelcome” has the meaning of “unwelcome in the sense that the employee
did not solicit or incite it”, and in the sense that “the employee regarded the conduct as undesirable
or offensive”.165 The same definition is applied in Australian case law.166

“Unwantedness” should not be confused with whether the participation was voluntary or not.167 The
fact that the victim eventually decides to submit or to endure the harassment, does not mean that
it is welcome.

According to the United States Equal Employment Opportunity Commission’s principles on


harassment168, making a contemporaneous complaint or protest will count in the complainant’s
favour when the court is presented with conflicting evidence in relation to the question whether the
conduct is unwanted. The relevance of a contemporaneous complaint, however, depends on the
nature of the sexual advances and the context in which the incidents occurred, such as fear of
retaliatory action, e.g. the loss of a job. It should not be a necessary element for the success of a
sexual harassment claim.169

Failure to respond positively to sexual advances can also indicate that the advances are
unwelcome. The Australian Sexual Harassment Code170 urges employers to remember that a
person may think that his/her advances are welcome, whilst the recipient only wants to avoid
confrontation. This is especially the case where there is a difference in age, racial or cultural
background, seniority, or personal power. However, the inconsistency of a complainant’s conduct
may adversely affect the credibility of his/her evidence regarding unwelcome conduct.

The fact that a complainant has welcomed conduct from a person does not imply that s/he would
subsequently welcome similar conduct from another or even the same person. In Swentek v US
Air Inc171, the court declined a view that it could be deduced from the plaintiff’s own past conduct
that she would welcome the conduct of her harasser. In County Fair Foods (Pty) Ltd v CCMA172
the victim and the harasser had previously been involved in a relationship, which was ended by
the complainant. Her former boyfriend tried to resume the relationship, forced her to sit on his lap
and struck her on the buttocks and thigh with a broomstick at work. The fact that they had been
romantically involved did not change the fact that he had harassed and assaulted her.

165 Henson v City of Dundee 682 F2d 897 (1982) at 903.


166 Alridge v Booth (1988) 80 ALR 1 at 5.
167 Equal Employment

168 Id n 7.

169 Id 4 - 5

170 Human Rights and Equal Opportunities Commission (Australia) op cit (1996) 17.

171 820 F2d 552, 557, 44 EPD 37, 457 (4 th Cir 1987).

172 [1999] 11 BLLR 1117 (LAC).


79

A controversial issue in this respect is the question whether the defendant/respondent may produce
evidence of the complainant’s sexual history or conduct in the workplace, e.g. the wearing of
sexually provocative clothes or participation in sexual discussions, etc. On the one hand, it is
argued that proof of prior sexual promiscuity by the complainant, or the wearing of sexually
provocative clothing in the workplace, may be indicative of susceptibility to sexual attentions. In
general the courts, however, are not keen to allow character evidence of this nature in sexual
harassment cases. The reason is that such evidence may be unnecessarily prejudicial to the
complainant and may discourage valid sexual harassment complaints. The fact that the
complainant has welcomed sexual attention from certain individuals in the past in given
circumstances, or wears clothes that some may find provocative, does not mean that s/he will find
the same attentions in other circumstances welcome. In Priest v Rotary (634 F Supp 571) the court
found in favour of a prostitute in a sexual harassment case.

(b) “Of a sexual nature”

This requirement refers to the nature of the conduct constituting harassment as sexual. It must not
be understood too narrowly. In one of the first South African sexual harassment cases, J v M173,
conduct constituting sexual harassment was said to range from innuendo, suggestions or hints, to
fondling without consent, force, and rape.

Item 4 of the Sexual Harassment Code contains a list of examples of sexual harassment, divided
into physical contact, verbal forms and non-verbal forms.

Physical contact includes touching, sexual assault and rape, and strip searches by or in the
presence of the opposite sex.

Verbal forms include, for example, innuendo’s suggestions, and comments with sexual overtones,
sex-related jokes, unwelcome graphic comments about a person’s body174 and whistling.

Non-verbal forms include gestures, indecent exposure and the unwelcome display of sexually
explicit pictures and objects, as well as letters and telephone calls of a sexual nature.

In view of electronic developments, sending or forwarding offensive or degrading e-mails, or using


the Internet at work to view offensive materials where others are also exposed to it, will constitute
sexual harassment as well. Office bantering and jokes “done in jest” may not always constitute
sexual harassment but will not constitute a valid defence in a sexual harassment claim if they are
indicative of underlying conceptions of prejudice and stereotyping that violate the dignity of those
against whom they are aimed. Exposure to offensive jokes and language, obscene gestures and
conversation and offensive posters, magazines and graffiti can all constitute sexual harassment.
According to item 4 of Sexual Harassment Code and comparative law, posters, graffiti, jokes, etc.,

173 (1988) 10 ILJ 755 at 757F.


174 In the United Kingdom a single incident where a manager greeted an employee with “Hiya, big tits” was
sufficient to constitute sexual harassment (Institu Cleaning Co Ltd v Heads [1995] IRLR 4 EAT). In Pretorius
v Britz [1997] 5 BLLR 649 (CCMA) complementing an employee on her legs was, together with numerous
other instances of sexual harassment, regarded as a continuation of the sexual harassment.
80

are now regarded as constituting forms of sexual harassment. Posters of nude women in
pornographic poses were declared a visual assault on female employees in Robinson v
Jacksonville Shipyards.175 The Court accepted expert evidence regarding the cumulative,
corrosive effects of the posters that run counter to the ideal of opening up the workplace to women.

(c) Persistency or seriousness

As was indicated above, it is usually considered a requirement for sexual harassment that the
impugned conduct must be persistent or serious. The fact that persistency or seriousness is
required, means that a single incident, if sufficiently serious, can constitute sexual harassment. It
is, therefore, not always necessary for the complainant to prove a pattern or a series of incidents
(J v M supra). In the United Kingdom case of Insitu Cleaning Co Ltd v Head176, it was held that
the single act of a manager greeting a female employee by saying “Hi ya, big tits”, was sufficient
to constitute sexual harassment.

Hostile environment sexual harassment will usually take the form of a persistent pattern of
behaviour, rather than a single incident. The more serious the harassment, however, the lesser the
degree of persistency needs to be for a successful complaint.

Office banter may be used here as an example of the principles. “Banter” is defined as “good-
humoured teasing or playful repartee…to tease or to mock gently”177, or “a playful friendly exchange
of remarks.178 Office banter, although it may be offensive or unwanted, is not necessarily
harassment. It will depend on the seriousness or persistency of such conduct. “Relentless” sexual
banter is included in the Australian definition of a hostile or sexually permeated working
environment.179 Employers in industries or businesses dominated by men must be particularly
careful because general sexual jokes, rough conversation or jokes may create a sexually hostile
environment.

(d) Impairment of dignity

Since harassment is a form of unfair discrimination, the violation of a harassed person’s dignity is
an important aspect of any successful harassment complaint. The judgments of the Constitutional
Court emphasise the fact that the impairment of human dignity is inherent to discriminatory
behaviour.

3. Racial, ethnic and national origin harassment

Currently, in South African law, there are no coherent and clear rules specifically formulated on
how to prevent and deal with cases of racial harassment. However, as alluded to earlier the

175 760 F Supp 1486 (1991).


176 (1995) IRLR 4.
177 Reader’s Digest Universal Dictionary (1987).

178 Oxford Advanced Learner’s Dictionary (1999).

179 Human Rights and Equal Opportunities Commission (Australia) op cit (1996) 4.
81

principles applicable to sexual harassment apply in a substantially similar manner to other forms
of harassment, including racial harassment. As indicated earlier, the Promotion of Equality and
Prevention of Unfair Discrimination Act180 contains only a general prohibition of harassment. It
includes in its list of unfair discrimination on grounds of race “the dissemination of any propaganda
or idea, which propounds the racial superiority or inferiority of any person…”181 Ethnic/racial
superiority and corresponding attitudes of stereotyping and prejudice often underpin racial or ethnic
harassment.

Until recently, racial and ethnic harassment in South Africa has been dealt with mostly along the
lines of “abuse or foul language”, which is a form of misconduct in the workplace.182 Apart from a
recent CCMA determination on the dismissal of an employee who distributed a racist e-mail at his
workplace183, there has not been any case of racial harassment.

EMPLOYMENT EQUITY ACT 55 OF 1998 AMENDED CODE OF GOOD PRACTICE ON THE


HANDLING OF SEXUAL HARASSMENT CASES IN THE WORKPLACE

1. Introduction

1.1 The objective of this code is to eliminate sexual harassment in the workplace.

1.2 This code provides appropriate procedures to deal with sexual harassment and prevent its
recurrence.

1.3 This code encourages and promotes the development and implementation of policies and
procedures that will lead to the creation of workplaces that are free of sexual harassment, where
employers and employees respect one another’s integrity and dignity, their privacy, and their right
to equity in the workplace.

2. Application of the code

2.1 Although this code applies to the working environment 1 as a guide to employers,
employees and applicants for employment, the perpetrators and victims of sexual harassment may
include:

180 Section 1.
181 Section 7(a).
182 Lebowa Platinum Mines Ltd v Hill 1998 (7) BLLR 666 LAC; (1998) 7 LAC 8.13.1. See also SACCAWU

obo Mabunza v Standard Bank (1998) 7 CCMA 8.13.2: <http://www.irnet.co.za.>


183 “CCMA backs Toyota’s firing of e-mailer” News Archives 22 November 2000 (http://www.irnet.co.za)
82

2.1.1 owners

2.1.2 employers

2.1.3 managers

2.1.4 supervisors

2.1.5 employees

2.1.6 job applicants

2.1.7 clients

2.1.8 suppliers

2.1.9 contractors

2.1.10 others having dealings with a business

2.2 Nothing in 2.1 above confers the authority or obligation on employers to take disciplinary
action in respect of non-employees.

2.3 A non-employee who is a victim of sexual harassment may lodge a grievance with the
employer of the harasser, where the harassment has taken place in the workplace or in the course
of the harasser’s employment.2

2.4 Where the term “employee” is used in this code, it will be deemed to include applicants for
employment.

(1) & (2) Where sexual harassment occurs outside of the working environment, regard
should be had to the Promotion of Equality and Prevention of Unfair Discrimination Act 4 of 2000.

3. Sexual Harassment as a form of unfair discrimination

Sexual harassment in the working environment is a form of unfair discrimination and is prohibited
on the grounds of sex and/or gender and/or sexual orientation.3

(3) Section 6 of the Employment Equity Act 55 of 1998 provides that no person may unfairly
discriminate, directly or indirectly against an employee in any employment policy or practice, on
one or more grounds, including race, gender, sex, pregnancy, marital status, family responsibility,
ethnic or social origin, colour, sexual orientation, age, disability, religion, HIV status, conscience,
belief, political opinion, culture, language and birth.

4. Test for Sexual Harassment

Sexual harassment is unwelcome conduct of a sexual nature that violates the rights of an employee
and constitutes a barrier to equity in the workplace, taking into account all of the following factors:

4.1 whether the harassment is on the prohibited grounds of sex and/or gender and/or sexual
orientation;

4.2 whether the sexual conduct was unwelcome;

4.3 the nature and extent of the sexual conduct; and


83

4.4 the impact of the sexual conduct on the employee

5. Factors to establish sexual harassment

5.1 Harassment on a prohibited ground

5.1.1 The grounds of discrimination to establish sexual harassment are sex, gender and sexual
orientation.

5.1.2 Same-sex harassment can amount to discrimination on the basis of sex, gender and sexual
orientation.

5.2 Unwelcome conduct

5.2.1 There are different ways in which an employee may indicate that sexual conduct is
unwelcome, including non-verbal conduct such as walking away or not responding to the
perpetrator.

5.2.2 Previous consensual participation in sexual conduct does not necessarily mean that the
conduct continues to be welcome.

5.2.3 Where a complainant has difficulty indicating to the perpetrator that the conduct is
unwelcome, such complainant may seek the assistance and intervention of another person such
as a co-worker, superior, counsellor, human resource official, family member or friend.

5.3 Nature and extent of the conduct

5.3.1 The unwelcome conduct must be of a sexual nature, and includes physical, verbal or non
verbal conduct.

5.3.1.1 Physical conduct of a sexual nature includes all unwelcome physical contact,
ranging from touching to sexual assault and rape, as well as strip search by or in the presence of
the opposite sex.

5.3.1.2 Verbal conduct includes unwelcome innuendos, suggestions, hints, sexual


advances, comments with sexual overtones, sex-related jokes or insults, graphic comments about
a person’s body made in their presence or to them, inappropriate enquiries about a person’s sex
life, whistling of a sexual nature and the sending by electronic means or otherwise of sexually
explicit text.

5.3.1.3 Non-verbal conduct includes unwelcome gestures, indecent exposure and the
display or sending by electronic means or otherwise of sexually explicit pictures or objects.

5.3.2 Sexual harassment may include, but is not limited to, victimization, quid pro quo harassment
and sexual favouritism.

5.3.2.1 Victimization occurs where an employee is victimized or intimidated for failing to


submit to sexual advances.

5.2.3.2 Quid pro quo harassment occurs where a person such as an owner, employer,
supervisor, member of management or co-employee, influences or attempts to influence an
employee’s employment circumstances (for example engagement, promotion, training, discipline,
dismissal, salary increments or other benefits) by coercing or attempting to coerce an employee to
84

surrender to sexual advances. This could include sexual favouritism, which occurs where a person
in authority in the workplace rewards only those who respond to his or her sexual advances.

5.3.3 A single incident of unwelcome sexual conduct may constitute sexual harassment.

5.4 Impact of the conduct

The conduct should constitute an impairment of the employee’s dignity, taking into account:

5.4.1 the circumstances of the employee; and

5.4.2 the respective positions of the employee and the perpetrator in the workplace.

6. Guiding principles

Employers should create and maintain a working environment in which the dignity of employees is
respected. A climate in the workplace should also be created and maintained in which
complainants of sexual harassment will not feel that their grievances are ignored or trivialized, or
fear reprisals. Implementing the following guidelines can assist in achieving these ends:

6.1 Employers/management and employees are required to refrain from committing acts of
sexual harassment.

6.2 All employers/management and employees have a role to play in contributing towards
creating and maintaining a working environment in which sexual harassment is unacceptable. They
should ensure that their standards of conduct do not cause offence and they should discourage
unacceptable behaviour on the part of others.

6.3 Employers/management should attempt to ensure that persons such as customers,


suppliers, job applicants and others who have dealings with the business are not subjected to
sexual harassment by the employer or its employees.

6.4 Employers/management should take appropriate action in accordance with this code where
instances of sexual harassment occur in the working environment.

7. Sexual Harassment Policies

7.1 Employers should, subject to any existing collective agreements and applicable statutory
provisions in respect of sexual harassment, adopt a sexual harassment policy, which should take
cognisance of and be guided by the provisions of this code.

7.2 The contents of sexual harassment policies should be communicated effectively to all
employees.

7.3 The adoption of a sexual harassment policy and the communication of the contents of the
policy to employees, should, amongst other factors, be taken into consideration in determining
whether the employer has discharged its obligations in accordance with the provisions of section
60(2) of the Employment Equity Act (EEA).

7.4 Sexual harassment policies should substantially comply with the provisions of this code and
include at least the following statements:
85

7.4.1 Sexual harassment is a form of unfair discrimination on the basis of sex and/or gender
and/or sexual orientation which infringes the rights of the complainant and constitutes a barrier to
equity in the workplace.

7.4.2 Sexual harassment in the workplace will not be permitted or condoned.

7.4.3 Complainants in sexual harassment matters have the right to follow the procedures in the
policy and appropriate action must be taken by the employer.

7.4.4 It will be a disciplinary offence to victimize or retaliate against an employee who in good
faith lodges a grievance of sexual harassment.

7.5 The procedures to be followed by a complainant of sexual harassment and by an employer


when sexual harassment has occurred, should be outlined in the policy.

8. Procedures

Employers should develop clear procedures to deal with sexual harassment. These procedures
should enable the resolution of problems in a sensitive, efficient and effective way.

8.1 Reporting sexual harassment

8.1.1 Section 60(1) of the EEA provides that conduct in contravention of the EEA must
immediately be brought to the attention of the employer.

8.1.2 In instances of sexual harassment, the word “immediately” shall mean, as soon as is
reasonably possible in the circumstances and without undue delay, taking into account the nature
of sexual harassment, including that it is a sensitive issue, that the complainant may fear reprisals
and the relative positions of the complainant and the alleged perpetrator in the workplace.

8.1.2 Sexual harassment may be brought to the attention of the employer by the complainant or
any other person aware of the sexual harassment, for example a friend, colleague or human
resources official acting on the request of the complainant, where the complainant has indicated
that she/he wishes the employer to be made aware of the conduct. However, where the sexual
harassment is of a particularly serious nature, the complainant should be encouraged to inform the
employer.

8.2 Obligations of the employer

When sexual harassment has been brought to the attention of the employer, the employer should:

8.2.1 consult all relevant parties;

8.2.2 take the necessary steps to address the complaint in accordance with this code and the
employer’s policy; and

8.2.3 take the necessary steps to eliminate the sexual harassment.

8.3 The steps to be taken by the employer on receipt of a complaint by a complainant, should
include but not be limited to the following:

8.3.1 advising the complainant of the informal and formal procedures available to deal with the
sexual harassment, as set out in items 8.5, 8.6 and 8.7 of this code;
86

8.3.2 where reasonably practicable, offering the complainant advice, assistance and counselling
as set out in item 8.4 of this code, including during any disciplinary enquiry that may be instituted;
and

8.3.3 following the procedures required by items 8.5, 8.6 and 8.7 of this code, in a manner that
is procedurally and substantively fair.

8.4 Advice and assistance

8.4.1 A complainant of sexual harassment may require advice and assistance, including
counselling.

8.4.2 As far as is practicable, employers should designate a person outside of line management
who complainants may approach for confidential advice and/or counselling. Such person:

8.4.2.1 could be a person employed by the employer to perform such a function, a trade
union representative, a co-employee or a professional engaged to perform such activity;

8.4.2.2 should have the appropriate skills and experience, including counselling and labour
relations skills; and

8.4.2.3 should be properly trained and given adequate resources.

8.5 Advising the complainant of workplace procedures to deal with sexual harassment

8.5.1 When an incident of sexual harassment is brought to the attention of an employer, such
employer should:

8.5.1.1 advise the complainant that there are formal and informal procedures which could
be followed to deal with the problem;

8.5.1.2 explain the formal and informal procedures to the complainant;

8.5.1.3 advise the complainant that she/he may choose which procedure should be
followed by the employer, except that in certain limited circumstances, as set out in clause 8.7.2,
the employer may choose to follow a formal procedure even if the complainant does not wish to do
so;

8.5.1.4 re-assure the complainant that she/he will not face job loss or any adverse
consequences if she/he chooses to follow either the formal or informal procedure;

8.5.1.5 advise the complainant that the matter will be dealt with confidentially if the
complainant so chooses.

8.6 Informal procedures

8.6.1 A complainant of sexual harassment may choose to follow either of the following informal
procedures:

8.6.1.1 the complainant or another appropriate person explains to the perpetrator that the
conduct in question is not welcome, that it offends the complainant, makes him or her feel
uncomfortable and that it interferes with his or her work; or
87

8.6.1.2 an appropriate person approaches the perpetrator, without revealing the identity of
the complainant, and explains to the perpetrator that certain forms of conduct constitute sexual
harassment, are offensive and unwelcome, make employees feel uncomfortable, and interfere with
their work.

8.6.2 An employer should consider any further steps, which can be taken to assist in dealing with
the complaint.

8.7 Formal procedure

8.7.1 A complainant may choose to follow a formal procedure, either with or without first following
an informal procedure.

8.7.2 In the event that a complainant chooses not to follow a formal procedure, the employer
should still assess the risk to other persons in the workplace where formal steps have not been
taken against the perpetrator. In assessing such risk the employer must take into account all
relevant factors, including the severity of the sexual harassment and whether the perpetrator has
a history of sexual harassment. If it appears to the employer after a proper investigation that there
is a significant risk of harm to other persons in the workplace, the employer may follow a formal
procedure, irrespective of the wishes of the complainant, and advise the complainant accordingly.

8.7.3 The employer’s sexual harassment policy and/or collective agreement should outline the
following in respect of a formal procedure:

8.7.3.1 with whom the employee should lodge a grievance;

8.7.3.2 the internal grievance procedures to be followed, including provision for the
complainant’s desired outcome of the procedures;

8.7.3.3 time frames which will allow the grievance to be dealt with expeditiously;

8.7.3.4 that should the matter not be satisfactorily resolved by the internal procedures
outlined above, a complainant of sexual harassment may refer the dispute to the Commission for
Conciliation Mediation and Arbitration (CCMA). Similarly an alleged perpetrator of sexual
harassment may refer a dispute arising from disciplinary action taken by the employer to the
CCMA; and

8.7.3.5 that it will be a disciplinary offence to victimize or retaliate against a complainant


who in good faith lodges a grievance of sexual harassment.

8.8 Disciplinary sanctions

The employer’s sexual harassment policy should specify the range of disciplinary sanctions that
may be imposed on a perpetrator. The sanctions must be proportionate to the seriousness of the
sexual harassment in question, and should provide that:

8.8.1 warnings may be issued for minor instances of sexual harassment;

8.8.2 dismissal may ensue for continued minor instances of sexual harassment after warnings,
as well as for serious instances of sexual harassment;

8.8.3 in appropriate circumstances upon being found guilty of sexual harassment, a perpetrator
may be transferred to another position in the workplace.
88

9. Confidentiality

9.1 Employers and employees must ensure that grievances about sexual harassment are
investigated and handled in a manner that ensures that the identities of the persons involved are
kept confidential.

9.2 In cases of sexual harassment, management, employees and the parties concerned must
endeavour to ensure confidentiality in the disciplinary inquiry. Only appropriate members of
management as well as the aggrieved person, representatives, alleged perpetrator, witnesses and
interpreter if required should be present in the disciplinary inquiry.

9.3 Employers are required to disclose to the complainant, the perpetrator and/or their
representatives, such information as may be reasonably necessary to enable the parties to prepare
for any proceedings in terms of this code.

10. Additional sick leave

10.1 Where an employee’s existing sick leave entitlement has been exhausted, the employer
should give due consideration to the granting of additional paid sick leave in cases of serious sexual
harassment, where the employee, on medical advice, requires trauma counselling.

10.2 In appropriate circumstances, employers may give consideration to assisting with the cost
of the medical advice and trauma counselling, where such amounts are not covered by any
applicable medical aid scheme.

11 Information and education

11.1 Where feasible, the Department of Labour should endeavour to ensure that copies of this
code are accessible and available in the official languages.

11.2 Employers and, where applicable, employer organizations should include the issue of
sexual harassment in their orientation, education and training programs.

11.3 Trade unions should include the issue of sexual harassment in their education and training
programs for shop stewards and employees.

11.4 CCMA commissioners should receive specialised training to deal with sexual harassment
cases.

UNIT 5

AFFIRMATIVE ACTION

See the extract from Pretorius JL, Klinck ME & Ngwena CG. 2001. Employment

Equity Law included in the study material.


89

UNIT 6

REASONABLE ACCOMMODATION

Pretorius et al: Employment Equity Law: Chapter 7.

UNIT 7

PREFERENTIAL PROCUREMENT AND BLACK ECONOMIC EMPOWERMENT

See study guide

You might also like