CONSTITUTIONAL
LITIGATION IN SOUTH AFRICA
Stes de Necker
Constitutional litigation in
South Africa is an area of the law dealing with the rules and
principles applicable to cases that involve constitutional matters. It examines
the constitutional jurisdiction of the Constitutional Court, the High Court and
the Supreme Court of Appeal (and certain other specialist courts), and
considers the various rules peculiar to these courts that are relevant to
constitutional litigation, such as the admission of an amicus curiae, the duty to
raise a constitutional matter as early as possible in the proceedings, and the
duty to join the relevant organ of state in a case involving a constitutional
issue.
Litigating
fundamental rights
The Bill of Rights is “the
principal source of substantive constraints on public power in the
Constitution.” The Bill of Rights instructs the state to use the power
that the Constitution gives it in ways that do not violate fundamental rights
and that promote and fulfil those rights. Should it fail to comply with these
instructions, it will act unconstitutionally; its actions or laws will be
unlawful and invalid. Although the Constitution as a whole is mostly concerned
with state power and with law, there are a number of provisions of the Bill of
Rights that place duties on private individuals in certain circumstances.
One of the most important
principles of South African law is expressed by the maxim ubi ius ubi remedium: Where there is a right, there is a remedy.
This means that the existence of
a legal rule implies the existence of an authority with the power to grant a
remedy if that rule is infringed. A legal rule will be deficient if there is no
means of enforcing it, and if no sanction attaches to a breach of that rule.
Litigation to enforce directly the Bill of Rights is one way in which breaches
of the Bill may be remedied.
Direct Bill of Rights litigation
is most conveniently analysed as taking place in distinct stages. There is an
initial procedural stage, followed by a number of stages in which issues of
substance are considered.
Initially, a court hearing a
Bill-of-Rights case will be concerned with any procedural issues that might
arise:
- the application of the Bill of
Rights to the subject-matter of the litigation;
- the justiciability of the issue
to be decided (including the standing of the applicant); and
- the jurisdiction of the court to
grant the relief claimed by the applicant.
Often, however, these procedural
aspects of the case will be uncontroversial and the court can proceed directly
to the substance of the case. The first step in the substantive stage of the
litigation involves interpreting the provisions in the Bill of Rights. The
court must consider, by reference to the facts of the case, and to the
interpretation of the Bill of Rights, whether a right has been violated. If the
court does find that a right has been violated, it must then consider whether
that violation is a justifiable limitation of the right. Finally, if the court
finds that a violation of a right is not a justifiable limitation, it will have
to consider the proper remedy to deal with the unconstitutional infringement of
a fundamental right. At each distinct stage of the litigation, the court must
consider whether the onus of proof is on the applicant or respondent.
Stages
Before a court may proceed to
consider the issues of substance raised in human rights litigation, it must
first consider a series of preliminary, procedural issues.
Procedural issues
Application and the principle of avoidance
Application of the Bill of Rights
concerns the issues of whether and how the Bill of Rights applies in a legal
dispute. The first of these issues (whether the Bill of Rights applies) raises
four questions:
- Who benefits from the Bill of
Rights?
- Who is bound by the Bill of
Rights?
- Does the Bill of Rights apply to
matters arising before its commencement?
- Does the Bill of Rights apply
only in the national territory or does it have extraterritorial effect?
Answering these questions
involves determining the “reach” or “scope” of the Bill of Rights.
Application is concerned not only
with whether the Bill of Rights applies, however, but also with how it applies
in a legal dispute. The question here is this: What is the relationship between
the Bill of Rights and the principles or rules of ordinary law? Currie and de
Waal argue that the two application issues should be resolved in the following
way:
The reach of the Bill of Rights
(beneficiaries, duties, time and territory) demarcates the types of legal
disputes to which the Bill of Rights directly applies. Within this area, the
Bill of Rights overrides ordinary law and conduct that is inconsistent with it.
In addition, and subject to considerations of justiciability and jurisdiction,
the Bill of Rights generates its own set of remedies. This form of application,
which is geared towards showing inconsistency between the Bill of Rights and
law or conduct, is called the direct application of the Bill of Rights.
At the same time, the Bill of
Rights contains a set of values that must be respected whenever the common law
or legislation is interpreted, developed or applied. This form of application,
which aims at creating harmony between the Bill of Rights and ordinary law, is
called the indirect application of the Bill of Rights. When indirectly applied,
the Bill of Rights does not override ordinary law; nor does it generate its own
remedies. Instead, law is interpreted or developed in a way that makes it conform
to the Constitution. The special constitutional rules relating to the
procedural issues of standing and the jurisdiction of the courts are also
irrelevant to this form of application. Rather, the Bill of Rights respects the
procedural rules, the purpose and remedies of ordinary law, but demands the
furtherance of its values through the operation of ordinary law.
Indirect application of the Bill
of Rights must be considered before direct application. This is the result of
the operation of the principle that constitutional issues should, where
possible, be avoided.
The principle of avoidance requires a court first to try
to resolve a dispute by applying ordinary legal principles, as interpreted or
developed with reference to the Bill of Rights, before applying the Bill of
Rights directly to the dispute.
An important implication of the
principle of avoidance is that the special rules in the Bill of Rights relating
to the standing of litigants and the jurisdiction of the courts apply only when
it is impossible to give effect to the values in the Bill of Rights by
applying, interpreting or developing the ordinary law. Similarly,
constitutional remedies are only relevant when the Bill of Rights is directly
applied to an issue. If it is possible to resolve the dispute through indirect
application, ordinary procedural rules and remedies apply to the dispute.
However, in order to apply the
Bill of Rights indirectly, a court must obviously determine the effect of the
provisions of the Bill of Rights. A court cannot interpret or develop the
ordinary law with reference to the values contained in the Bill of Rights
without knowing what they are. The interpretation of the Bill of Rights and its
limitation clause therefore remain important, even when the Bill of Rights is indirectly
applied.
Justiciability
In some cases, an applicant may
lack standing to seek a remedy. In other cases, the issue may have become moot
or academic and therefore not justiciable. An issue may also be non-justiciable
because it is not yet ripe for decision by a court. The Bill of Rights contains
special rules relating to these issues when it is directly applied. In such
cases, it demands a broader approach to standing. In cases of indirect
application, the ordinary legal rules apply.
Jurisdiction
In cases of indirect application,
the ordinary procedural rules apply. The constitutional jurisdiction of the
courts and the procedures that must be observed when the Bill of Rights is
directly applied to law or conduct are not simply technical issues, but are of
paramount importance for the protection of fundamental rights in practice. It
is important to know in which forum to challenge an alleged violation of a
right, since not all courts have the same jurisdiction in constitutional
matters. If the court does not have the jurisdiction to grant the relief
claimed, it must dismiss the application, whatever its merits may be.
Substantive questions
At the substantive stage of
Bill-of-Rights litigation, the court is concerned with the substance of the
applicant’s allegation that a right has been infringed by law, or by the
conduct of the other party. The court must assess the merits of this
allegation. This assessment primarily involves the interpretation of the
provisions of the Constitution in general, and the Bill of Rights in
particular.
Interpretation
The court must determine whether
or not the Bill of Rights protects a particular interest of the applicant. It
must then determine whether or not the law that has been challenged or the
conduct of the respondent impairs that interest, thereby trespassing in an area
protected by the Bill of Rights.
Limitation
Fundamental rights and freedoms
are not absolute: “Their boundaries are set by the rights of others and by the
legitimate needs of society. Generally, it is recognised that public order,
safety, health and democratic values justify the imposition of restrictions on
the exercise of fundamental rights.” In the South African Constitution, a
general limitation clause, section 36, sets out specific criteria for the restriction
of the fundamental rights in the Bill of Rights. The clause is general because
it applies in the same way to all the rights in the Bill of Rights. In this
regard the Constitution differs from, for example, the United States
Constitution, which does not contain a limitations clause at all. The German
Bill of Rights does not have a general limitations clause, but it does contain
specific limitations clauses attached to most of the fundamental rights.
The court, in determining whether
or not a right has been unjustifiably infringed, undertakes what is essentially
a two-stage exercise. First, there is the threshold enquiry aimed at
determining whether or not the enactment in question constitutes a limitation
on one or other guaranteed right. This entails examining (a) the content and
scope of the relevant protected right(s) and (b) the meaning and effect of the
impugned enactment to see whether there is any limitation of (a) by (b).
Subsections (1) and (2) of section 39 of the Constitution give guidance as to
the interpretation of both the rights and the enactment, essentially requiring
them to be interpreted so as to promote the value system of an open and
democratic society based on human dignity, equality and freedom. If upon such
analysis no limitation is found, that is the end of the matter. The
constitutional challenge is dismissed there and then.
If, however, the court determines
that a law, or the conduct of the respondent, impairs a fundamental right, “the
second stage ensues. This is ordinarily called the limitations
exercise.” It must then consider whether the infringement is nevertheless
a justifiable limitation of the right in question: “In essence this requires a
weighing-up of the nature and importance of the right(s) that are limited together
with the extent of the limitation as against the importance and purpose of the
limiting enactment.”
Not all laws or conduct that
infringe fundamental rights are unconstitutional. Sometimes a law may be a
justifiable limitation on a particular fundamental right. This means that,
although the law or conduct infringes the right, the infringement (which is
called a limitation) is justifiable. While infringing conduct itself cannot
validly limit a fundamental right, the challenged conduct may be authorised by
law. If the law passes the limitations test, the conduct it authorises will
survive a constitutional challenge. The question, ultimately, which falls for
determination by this balancing exercise is whether or not a balance can be
struck? If not—if, that is, there is no proportionality—the limitation will not
be found to be reasonable and justifiable in an open and democratic society.
In the case of the death penalty,
for example, the purposes served by such a penalty—at least, the purposes as
considered by the Constitutional Court in S v Makwanya—are deterrence, prevention. and
retribution. In serving these purposes, however, the impugned right is
obliterated completely; its essential content, according to the Constitutional
Court, is negated. There is, in other words, no proportionality between
means and end.
One consequence of the inclusion
of a general limitation clause in the Bill of Rights is that the process of
considering the limitation of fundamental rights must be distinguished from
that of interpretation of the rights. If it is argued that conduct or a
provision of the law infringes a right in the Bill of Rights, it will first
have to be determined whether that right has in fact been infringed, and
thereafter whether the infringement is justified. The question of whether an
infringement of a right is a legitimate limitation of that right “frequently
involves a far more factual enquiry than the question of interpretation.” Appropriate
evidence must be led to justify the limitation of a right in accordance with
the criteria laid down in section 36. A court cannot determine in the abstract
whether the limitation of a right is “reasonable” or “justifiable in an open
and democratic society based on human dignity, equality and freedom.” This
determination requires “evidence, such as sociological or statistical data,” on
the impact that the legislative restriction has on society.
Remedies
Should a court find that a right
has been infringed, and that the infringement does not satisfy the test for a
valid limitation of a right, the question of the appropriate remedy for the
infringement arises. The constitutional remedies are only available when the
Bill of Rights is directly applied. In cases of indirect application, ordinary
legal remedies are used to give effect to the fundamental values in the Bill of
Rights.
Onus
The Constitutional Court has
dealt with the issue of onus, or burden of proof, by dividing the substantive
stage of human-rights litigation into two further sub-stages:
interpretation; and limitation.
The approach of the court to onus
in respect of these stages is set out by Ackermann J in the following extract
from Ferreira v Levin NO:
The task of determining whether
the provisions of [an] Act are invalid because they are inconsistent with the
guaranteed rights here under discussion involves two stages, first, an enquiry
as to whether there has been an infringement of the guaranteed right; if so, a further enquiry as
to whether such infringement is justified under the limitation clause. The task of
interpreting the fundamental rights
rests, of course, with the Courts, but it is for the applicants to prove the facts
upon which they rely for the claim of infringement of the particular right in
question. Concerning the second stage, [it] is for the legislature or the party
relying on the legislation to establish this justification [in terms of the
limitation clause], and not for the party challenging it, to show that it was
not justified.
This description focuses on the
difference in onus relating to the questions of interpretation and limitation.
The applicant has to show that an infringement of a right has taken place. This
requires the applicant to prove the facts on which he relies. The respondent
then has to show that an infringement is a justifiable limitation of the right
in terms of section 36.
In addition to the applicant’s
onus at the substantive stage of litigation, the applicant must also show, at
the preliminary stage of litigation,
- that the Bill of Rights applies
to the challenged law or conduct;
- that the issue is justiciable;
- that he or she has standing; and
- that he or she is in the right
forum to obtain the desired relief.
Only once these issues have been
decided in the applicant’ s favour, and a violation of the Bill of Rights is
found, will the party relying on the validity of the challenged decision or
legislation be called upon to justify it in terms of section 36, the limitation
clause.
The question of who bears the
onus when considering the appropriate relief for unconstitutional legislation
or conduct is “more complicated.” When the Bill of Rights is indirectly
applied, an ordinary legal remedy is granted, and the ordinary legal rules
apply in respect of the burden of proof.
When the Bill of Rights is directly
applied, the remedy that flows from a finding of inconsistency between the Bill
of Rights, on the one hand, and law or conduct, on the other, is invalidation
by the court of the offending law or conduct. A party proposing a variation of
this form of relief in terms of section 172(l)(b)(i) or (ii) must justify the
request. Since section 172 allows a court to limit or suspend the effects of a
declaration of invalidity, the respondent will in most cases be called upon to
justify such a request. However, a court may also grant relief in addition to
the declaration of invalidity, as in the case of an interdict or constitutional
damages. In most cases, the applicant will request such relief, and will
therefore bear the burden of persuasion.
Summary
The stages through which
Bill-of-Rights litigation typically proceeds may be summarised as follows:
Procedural stage
Does the Bill of Rights apply in
the dispute between the parties? How does the Bill of Rights apply in the
dispute?
As for the first question, it
must be determined whether the applicant is entitled to claim the benefits of
the Bill of Rights. It must also be determined whether the Bill of Rights
applies to the conduct of the respondent, in the sense that the respondent has
obligations under the Bill of Rights. Finally, it must be determined whether
the cause of action arose in the national territory during the period of
application of either the interim or 1996 Bill of Rights.
As for the second question,
indirect application must be considered before direct application. If the Bill
of Rights is indirectly applied to the issue, the questions of justiciability,
jurisdiction and an appropriate remedy are resolved in terms of ordinary legal
rules. If directly applied, special constitutional rules apply.
Is the issue to be decided
justiciable?
Does the applicant in the matter have standing in respect of the
particular relief sought?
Does the court have jurisdiction
to grant the relief claimed?
Substantive stage
If the answer to all three
questions is “yes,” the court is able to move on to the substantive stage.
Has the law or conduct of the
respondent infringed a fundamental right of the applicant? If so, the court
will consider whether the infringement is justifiable. If not, then the
application must be dismissed.
Is the infringement a justifiable
limitation of the right in question according to the criteria set out in
section 36? If yes, then the conduct of the respondent is not unconstitutional;
the application must be dismissed. If no, then the conduct is unconstitutional,
and the question of the appropriate remedy must be canvassed.
Remedies
What remedy is appropriate in
this case?
Application of the Bill of Rights
The Bill of Rights applies
directly to a legal dispute when a right of a beneficiary of the Bill of Rights
has been infringed by a person or entity on whom the Bill of Rights has imposed
the duty not to infringe the right during the period of operation of the Bill
of Rights in the national territory.
In addition, in instances when the
Bill of Rights does not apply directly to a dispute, because one or more of the
elements above is not present, it may apply indirectly. This is because all law
must be developed, interpreted and applied in a way that conforms to the Bill
of Rights.
A conceptual distinction, then,
must be made between two forms of application of the Bill of Rights:
Indirect application: The
Constitution and the Bill of Rights establish an “objective normative value
system,” a set of values that must be respected whenever the common law or
legislation is interpreted, developed or applied. This form of application is
termed the “indirect” application of the Bill of Rights. When indirectly
applied, the Bill of Rights does not override ordinary law or generate its own
remedies. Rather, the Bill of Rights respects the rules and remedies of
ordinary law, but demands furtherance of its values mediated through the
operation of ordinary law.
Direct application: In disputes
in which the Bill of Rights applies as directly applicable law, it overrides
ordinary law and any conduct that is inconsistent with it. To the extent that
ordinary legal remedies are inadequate or do not give proper effect to the
fundamental rights, the Bill of Rights generates its own remedies. The
methodology for the conduct of direct-rights litigation is applicable.
This distinction was of “decisive
significance” under the interim Constitution. It has “comparatively less
significance” under the 1996 Constitution. This is because of changes made
to the jurisdictional and application scheme by the 1996 Constitution. There
are nevertheless important consequences that follow from the form of
application, and the distinction between direct and indirect application
therefore “continues to play a role in constitutional litigation.”
The application of the Bill of
Rights has been “one of the most troublesome issues in South African
constitutional law.” The principal reason for the difficulty is that,
since 1994, South Africa has had two Constitutions that have treated the issue
differently. Much of the relevant jurisprudence, particularly relating to the
application of the Bill of Rights to the common law, was decided under the
interim Constitution and does not always bear precisely on the altered
jurisdictional and application schemes of the 1996 Constitution.
Interim Constitution
The narrowest conception of a
bill of rights is that it is a “charter of negative liberties.” This means that
it is intended to protect individuals against state power by listing rights
that cannot be violated by the state, either by means of law or through the
conduct of state actors. This is the “vertical” relationship—between
individuals and the state. A bill of rights that has solely vertical
application will place duties on the state not to violate the rights of
individuals. It will not place any similar duties on individuals.
According to the Constitutional
Court, in Du Plessis v De Klerk, the Bill of Rights in Chapter 3 of the interim
Constitution conformed to this traditional model, in so far as it had no direct
application to so-called “horizontal” disputes: that is, to disputes between
private litigants governed by the common law. “Constitutional rights under
Chapter 3,” the court held, “may be invoked against an organ of government but
not by one private litigant against another.” This was principally because
of the absence of the word “judiciary” in section 7, the application section of
the interim Constitution, which provided that the Bill of Rights “shall bind
all legislative and executive organs of state at all levels of government.” The
omission meant that the Bill of Rights placed duties to uphold constitutional
rights only on the legislative and executive organs of state. Individuals were
not directly bound by the Bill of Rights. Nor was the judiciary, which had the
task of adjudicating and enforcing the rights and duties of individuals.
However, while the interim Bill
of Rights did not apply directly to horizontal cases, it did have indirect
application. The Bill of Rights applied to “all law in force,” including all
pre- and post-1994 legislation and the uncodified common law (provisions of the
common law which had not been incorporated into legislation). Even if
individuals were not directly bound by the Bill of Rights, the courts had to
interpret legislation, and develop the common law, so that the ordinary law
recognised and protected the rights in the Bill of Rights. In Du Plessis v
Klerk, the Constitutional Court held that the Bill of Rights in the Interim
Constitution “may and should have an influence on the development of the common
law as it governs relations between individuals.” This was provided for in
section 35(3) of the interim Constitution: “In the interpretation of any law and
the application and development of the common law and customary law, a court
shall have due regard to the spirit, purport and objects of” the Bill of
Rights. “In private litigation,” held the court, “any litigant may nonetheless
contend that a statute (or executive act) relied on by the other party is
invalid as being inconsistent with the limitations placed on legislature and
executive under Chapter 3.” Accordingly, “as Chapter 3 applies to common
law, governmental acts or omissions in reliance on the common law may be
attacked by a private litigant as being inconsistent with Chapter 3 in any
dispute with an organ of government.”
In Du Plessis, the Constitutional
Court also decided a crucial jurisdictional issue. The court’s conclusion that
the Constitution distinguished between direct and indirect application of the
Bill of Rights was bolstered by the close fit between this distinction and the
“two-track” jurisdictional scheme of the interim Constitution, which
distinguished between “constitutional matters” and other matters, the former
being the preserve of the Constitutional Court, and the latter the preserve of
the Appellate Division. The development of the common law was a
non-constitutional matter, and therefore remained within the jurisdiction of
the court that had overseen the development of the common law for the past
century: the Appellate Division of the Supreme Court. “The development of the
common law,” held the court in Du Plessis, “is within the jurisdiction of the
Appellate Division, but not of the Constitutional Court.”
1996 Constitution
With the Du Plessis decision in
mind, and concerned that confining the Bill of Rights to direct vertical
application amounted to the toleration of private violations of rights, the
Constitutional Assembly created a different application and jurisdictional
scheme in the 1996 Constitution. To provide for direct horizontal application,
two textual changes were made. The first was the addition of the word
“judiciary” in section 8(1), missing from the application provisions of the interim
Constitution. The second was the imposition of a duty on individuals, in
section 8(2), to uphold the rights of other individuals: “A provision of the
Bill of Rights binds a natural or a juristic person if, and to the extent that,
it is applicable, taking into account the nature of the right and the nature of
any duty imposed by the right.”
The 1996 Constitution also made
significant changes to the powers of the courts to enforce the Constitution.
The “two-track” jurisdictional scheme of the interim Constitution was replaced
by a unified scheme in which the High Courts, Supreme Court of Appeal and the
Constitutional Court shared jurisdiction over constitutional matters. This
scheme required revision of the holding in Du Plessis that the application of the
Constitution to the common law was a non-constitutional matter. Under the 1996
Constitution, the Constitutional Court held in the Pharmaceutical Manufacturers
case, “there are not two systems of law, each dealing with the same subject
matter, each having similar requirements, each operating in its own field with
its own highest court. There is only one system of law. It is shaped by the
Constitution which is the supreme law, and all law, including the common law,
derives its force from the Constitution and is subject to constitutional
control.”
While clearly envisaging direct
horizontal application in applicable cases, the 1996 Bill of Rights also
requires the courts to apply the Bill of Rights indirectly, in similar terms to
section 35(3) of the interim Constitution. This is section 39(2): “When
interpreting any legislation, and when developing the common law or customary
law, every court, tribunal or forum must promote the spirit, purport and
objects of the Bill of Rights.”
To summarise, the 1996 Constitution,
like its predecessor, distinguishes two forms of application of the Bill of
Rights:
Direct application entails the
imposition of duties by the Bill of Rights on specified actors: A breach of
such a duty is a violation of a constitutional right.
Indirect application occurs where
there is a provision of ordinary law (legislation, common law or customary law)
that mediates between the Bill of Rights and the actors who are subject to that
law.
The duty of the courts is to ensure that the ordinary law conforms to the
values to which the Bill of Rights, by conferring the rights and duties that it
does, gives effect.
Like its predecessor, the 1996
Constitution provides for direct vertical application of the Bill of Rights
but, unlike its predecessor, does not confine itself to this form of direct
application. Section 8(2) clearly envisages direct application of the Bill of
Rights in the horizontal relationship in certain circumstances.
Direct application
There are four elements that are
determinative of the direct application of the Bill of Rights. The first
relates to beneficiaries, the second to the duties imposed by the Bill of
Rights, the third to time and the fourth to the limited territorial effect of
the Bill of Rights.
Legal rights are a correlative
relationship. If Armand has a legal right to something, this postulates that
Theo has a legal duty to Armand to uphold that right. Armand is therefore the
beneficiary of the right and Theo is the duty-bearer in respect of the right.
The first application issue to confront when considering the reach of the Bill
of Rights is to identify the beneficiaries and the duty-bearers of the rights
in the Bill of Rights.
Beneficiaries
Natural persons
Most of the rights in the Bill of
Rights are for the benefit of “everyone.” The negative phrase, which is to the
same effect, is that a right may be denied to “no-one.” For example, section 11
provides that “everyone has the right to life.” Section 13 is phrased
negatively but, like section 11, accords the right universally: “No one may be
subjected to slavery, servitude or forced labour.” Rights phrased in this way
are accorded to all natural persons within the territory of the Republic.
Other rights are accorded to
narrower categories of beneficiaries. The political rights in section 19, the
citizens’ rights in section 20, certain of the freedom-of-movement rights in
section 21, and the freedom-of-trade right in section 22, are accorded to
“every citizen.” The right to vote and to stand for political office, in
section 19(3), is restricted to “every adult citizen.” Further examples of
restrictions on the category of beneficiaries are the cultural rights contained
in section 31, which are for the benefit only of “persons belonging to a
cultural, religious or linguistic community.” The rights contained in section
35 are restricted to arrested, detained and accused persons.
The restriction of a right to a
particular category of beneficiaries is an attempt to circumscribe the scope of
the right. A right accorded only to citizens obviously has a more limited scope
of operation than a right accorded universally. The circumscription of rights
in this manner “does not really concern the application of the rights, but may
raise difficult issues of interpretation.” The courts will have to
interpret the Bill of Rights to determine who is, for example, a “detained
person,” or “a worker,” or a “person belonging to a cultural religious or
linguistic community.” The activities of persons who are excluded from the
scope of a right will not be protected by the right.
Juristic persons
Are the rights accorded to
“everyone” also available for the benefit of juristic persons? In other words,
are companies protected by the Bill of Rights? What about state-owned or
state-controlled corporations such as Eskom or the SABC? These questions are
answered by reference to s 8(4): “A juristic person is entitled to the rights
in the Bill of Rights to the extent required by the nature of the rights and
the nature of that juristic person.”
In order to decide whether a juristic
person is protected, regard must be had to two factors:
the nature of the fundamental
right in question; and the nature of the juristic person.
In Ex Parte Chairperson of the
Constitutional Assembly: In Re Certification of the Constitution of the Republic
of South Africa, an objection was raised that, inconsistently with
Constitutional Principle II, the extension of the rights guaranteed by the Bill
of Rights to juristic persons would diminish the rights of natural persons.
This Constitutional Court rejected the objection in the following terms:
Many “universally accepted
fundamental rights” will be fully recognised only if afforded to juristic persons
as well as natural persons. For example, freedom of speech, to be given proper
effect, must be afforded to the media, which are often owned or controlled by
juristic persons. While it is true that some rights are not appropriate to
enjoyment by juristic persons, the text of s 8(4) specifically recognises this.
The text also recognises that the nature of a juristic person may be taken into
account by a court in determining whether a particular right is available to
such person or not.
The nature of some of the
fundamental rights prevents them from benefiting juristic persons. The rights
to life and physical integrity, and to human dignity, for example, cannot
sensibly be applied to juristic persons. A company cannot claim protection of
its right to life or human dignity, or its right not to be deprived of liberty
or tortured, “because these rights protect aspects of human existence that a
company does not possess.” However, the nature of most of the rights that
are likely to be relied on by juristic personsmakes them applicable to the
protection of juristic persons. However, in the case of rights that stem from
the protection of human dignity (such as privacy), the Constitutional Court has
indicated, in Investigating Directorate: Serious Economic Offences and Others v
Hyundai Motor Distributors (Pty) Ltd In re: Hyundai Motor Distributors (Pty)
Ltd v Smit NO, that juristic
persons are entitled only to a reduced level of protection compared to natural
persons:
Privacy is a right which becomes
more intense the closer it moves to the intimate personal sphere of the life of
human beings, and less intense as it moves away from that core. This understanding
of the right flows from the value placed on human dignity by the Constitution.
Juristic persons are not the bearers of human dignity. Their privacy rights,
therefore, can never be as intense as those of human beings. However, this does
not mean that juristic persons are not protected by the right to privacy.
Exclusion of juristic persons would lead to the possibility of grave violations
of privacy in our society, with serious implications for the conduct of
affairs. The state might, for instance, have free licence to search and seize
material from any non-profit organisation or corporate entity at will. This
would obviously lead to grave disruptions and would undermine the very fabric of
our democratic state. Juristic persons therefore do enjoy the right to privacy,
although not to the same extent as natural persons.
It is the second of section
8(4)’s criteria (the nature of the juristic person) that may place greater
restrictions on the availability of human rights to juristic persons. “It is
difficult to see,” write Currie and De Waal,
how organs of state exercising
core government functions such as Parliament, a cabinet minister or the police
will ever be able to rely on the protection of the Bill of Rights. Although
arguably they are ‘juristic persons’, the nature of such organs of state makes
them unsuitable to be beneficiaries of fundamental rights. They are not used by
individuals for the collective exercise of their fundamental rights, but are
instead used by the state for the exercise of its powers.”
However, state-owned
corporations, such as the South African Broadcasting Corporation or the Post
Office, or entities such as universities, which are set up by the state for the
purpose, amongst other things, of realising particular fundamental rights, are
differently situated: “Clearly a state-owned corporation like the SABC should
be able to invoke the right to freedom of speech and the press when it becomes
involved in a dispute with the state or even with an individual.” In
Hoffmann v South African Airways, the Constitutional Court held,
Transnet is a statutory body,
under the control of the state, which has public powers and performs public
functions in the public interest. It was common cause that SAA is a business
unit of Transnet. As such, it is an organ of state and is bound by the
provisions of the Bill of Rights in terms of section 8(1), read with section
239, of the Constitution. It is, therefore, expressly prohibited from
discriminating unfairly.
As for private juristic persons,
the size or activities of the juristic person are not necessarily decisive.
Of
greater significance, in the view of Currie and De Waal, is “the relationship
between the activities of the juristic person and the fundamental rights of the
natural persons who stand behind the juristic person.” In other words,
juristic persons are not in and of themselves worthy of protection; they become
so when they are used by natural persons for the collective exercise of their
fundamental rights. For example, companies are routinely used by individuals as
an entity for conducting business, necessitating the exercise of property
rights by companies. As the Constitutional Court put it in First National Bank
of SA Limited t/a Wesbank v Commissioner for the South African Revenue Services.
It is trite that a company is a
legal entity altogether separate and distinct from its members, that its
continued existence is independent of the continued existence of its members,
and that its assets are its exclusive property. Nevertheless, a shareholder in
a company has a financial interest in the dividends paid by the company and in
its success or failure because she “... is entitled to an aliquot share in the
distribution of the surplus assets when the company is wound up” No matter how
complex the holding structure of a company or groups of companies may be,
ultimately—in the vast majority of cases—the holders of shares are natural
persons.
The court also commented on “the
universal phenomenon” that “natural persons are increasingly forming companies
and purchasing shares in companies for a wide variety of legitimate purposes,
including earning a livelihood, making investments and for structuring a
pension scheme.” The use of companies, the court found, “has come to be
regarded as indispensable for the conduct of business, whether large or small.
It is in today’s world difficult to conceive of meaningful business activity
without the institution and utilisation of companies.” The court saw that
“denying companies entitlement to property rights,” even more so than in
relation to the right to privacy, would
“lead to grave disruptions and
would undermine the very fabric of our democratic State.” It would have a
disastrous impact on the business world generally, on creditors of companies
and, more especially, on shareholders in companies. The property rights of
natural persons can only be fully and properly realised if such rights are
afforded to companies as well as to natural persons.
What section 8(4) envisages is
that there should be a link between protecting the activity of the juristic
person and protecting the fundamental rights of the natural persons that lie
behind it.
Much of the debate about the
meaning of the guidelines contained in section 8(4)—that is, “the nature of the
right” and “the nature of the juristic person”—is made “irrelevant,” write
Currie and De Waal, by the courts’ approach to standing in constitutional
litigation. A person has standing to
challenge the constitutionality of laws or conduct, provided that he alleges that a
fundamental right is infringed or threatened; and provided that he has, in terms of
the categories listed in section 38, a sufficient interest in obtaining a
remedy.
The first enquiry is objective:
It is sufficient to show that a right in the Bill of Rights is violated by a
law or conduct; it is not necessary to show that a right of the applicant has
been violated. This approach allows anyone with a sufficient interest to rely
on the objective inconsistency between the Bill of Rights and a law or conduct.
For example, it will seldom be necessary for juristic persons to invoke section
8(4), which sometimes extends the protection of the right to the juristic
person itself. Laws, and many forms of state and private conduct, inevitably
impact on the activities of both natural and juristic persons. Provided that a
juristic person has a sufficient interest of its own, or, if it is an
association, a sufficient interest of its members, it may challenge such laws
or conduct on the basis of fundamental rights that do not necessarily benefit
the juristic person. For example, a law which prohibits the sale of wine on
Sunday may be challenged by a company on the basis of the right to freedom of
religion, provided that the company has a sufficient interest in the outcome of
the litigation. It is not necessary in such a case for the company to show that
the right to freedom of religion benefits juristic persons.
It is only when a law or conduct
impacts solely on the activities of juristic persons that it will not be
possible to follow this course of action. Then there can be no objective
inconsistency between the Bill of Rights and the law or conduct, unless section
8(4) extends protection of the relevant right to juristic persons. For example,
when a special tax on companies is challenged, a person challenging the tax
will have to do so on the basis of a right that benefits juristic persons.
Waiver
Waiver may be considered an
application issue, and can be accommodated under the consideration of the
beneficiaries of the Bill of Rights in that someone who has waived a right has
agreed that he will not claim the benefit of it. Although the distinction may
be difficult to make in some cases, the waiver of fundamental rights should be
distinguished from a decision not to exercise a fundamental right. Where a
person chooses not to take part in an assembly, or not to join an association,
he cannot later complain about a violation of his rights to freedom of assembly
or association. The same applies when an arrested person makes an informed
choice to co-operate with the police by making a statement or a confession, or
when a person allows the police to search his or her home: “Such a person
cannot subsequently object at the trial that the introduction of the evidence
violates his or her right to remain silent or his or her right to privacy of
the home.” In principle, the
accused may nevertheless object to the use of the evidence if it would render
the trial unfair. In the absence, however, of other circumstances—for example,
that the accused was improperly persuaded to co-operate—“it is difficult to see
why the use of the evidence would result in an unfair trial.”
Waiver is different. One is
dealing with waiver when someone undertakes not to exercise a fundamental right
in future. For example, a contractual restraint of trade is an undertaking to
waive one’s right, guaranteed by section 22, to occupational freedom for a
period of time. A person may also undertake not to disclose sensitive
information, or undertake to vote for a particular political party on election
day; he may agree to have his telephone calls recorded and listened to by his
or her employer, or to attend religious instruction classes in a private
school. These are, respectively, attempts to waive the rights to freedom of
expression, to vote, to privacy and to freedom of religion.
The question is
whether people may be obliged to honour such an undertaking even if they
subsequently change their minds.
A waiver, write Currie and De
Waal, “cannot make otherwise unconstitutional laws or conduct constitutional
and valid.” Section 2 of the Constitution provides that law or conduct
inconsistent with the Constitution is invalid. This is an objective
consideration: “The actions of the beneficiary of the right can have no
influence on the invalidity of unconstitutional law or conduct.” That is
why a person cannot validly undertake to behave unconstitutionally; such an
undertaking will have no force and effect.
Similarly, a person cannot waive
the indirect application of the Bill of Rights. Two people may not undertake,
for example, that the law of defamation must be applied in future disputes
between them without any reference to the Bill of Rights. The reason for this
is that section 39(2) requires the courts to promote the Bill of Rights when
developing the common law. Individuals may not prevent the court from
fulfilling its constitutional obligations.
What individuals may do is to
waive the right to exercise a fundamental right. The individual may undertake
not to invoke the constitutional invalidity of state or private conduct.
Although, “from a constitutional point of view, such a waiver is hardly ever
decisive of an issue,” nonetheless “it is also seldom irrelevant.”Although waiver
is dealt with here as an issue of application, “we do not mean to suggest that
it must be answered by simply asking whether the individual may exclude him or
herself from the ‘benefits’ of a particular fundamental right in the
circumstances of the case. Waiver, and more
generally, victim responsibility, may also influence the limitation stage and
the remedy that a court will award for breach of the fundamental right.
The effect of waiver depends
firstly on the nature and purpose of the fundamental right in question. In
principle, many of the freedom rights may be waived as long as the undertaking
is made clearly and freely, and without the subject’s being placed under duress
or labouring under a misapprehension: “To be enforceable, however, it would
have to be a fully informed consent and one clearly showing that the applicant
was aware of the exact nature and extent of the rights being waived in
consequence of such consent.”
For example, the right to occupational
freedom is often waived by employees when concluding a contract of
employment. There is also no reason why one cannot waive the right to privacy. Similarly,
one may validly undertake not to demonstrate, not to join a political
party, or not to leave the Republic. One may also waive many of the
procedural rights, such as the right to legal representation or the right of
access to court. As far as these rights are concerned, it is not so much the
nature of the right that may be decisive, “but the length of the period of the
waiver, the danger of abuse and the position of the beneficiary.” The waiver
may not be contrary to some other constitutional principle or otherwise contra
bonos mores.
In contrast to the freedom
rights, the nature of the rights to human dignity, to life,
and
not to be discriminated against, or the right to a fair trial, does not
permit them to be waived: “Unlike the freedom rights, these rights cannot be
exercised negatively.” The right to freedom of expression, for example,
can be exercised by keeping quiet, but the right to dignity cannot be exercised
by being abused. One cannot therefore assume that the right is exercised when
it is waived (as one can, subject to the above considerations, with the freedom
rights).
Although some rights may not be
waived, this does not mean that the fact of waiver then becomes legally
irrelevant. Waiver may also be relevant when considering the remedy to be
awarded for the violation of a fundamental right. For example, a court would
not enforce an undertaking to vote for a particular political party, but it
would also probably not grant relief for such a violation of the right to vote,
other than to declare the agreement to be invalid. On the other hand, if a
person is prevented from voting against his or her will, “it may well be
appropriate to award damages for the infringement.”
“An interesting illustration” of
some of the principles discussed above is provided by Garden Cities Inc
Association v Northpine Islamic Society. The High Court granted an interdict
enforcing a contractual undertaking not to use loud-speaking equipment to broadcast
calls to prayer from a suburban mosque. An undertaking not to use any
amplification equipment had been given by the respondent in 1986, in the deed
of the sale of the land on which the mosque was built. Despite the contract,
the respondent started broadcasting amplified calls to prayer through a
loudspeaker, and the applicants applied for an interdict to stop it.
The argument of the respondent
was that enforcing the contract would amount to a violation of the
constitutional right to freedom of religion, and that the Constitution did not
permit the waiver of a fundamental aspect of one’s religion. Conradie J was
able to avoid the waiver issue by holding that amplification of the call to
prayer had not been shown to be a fundamental precept of the Islamic faith, and
that the agreement therefore did not infringe the right to religious freedom.
Currie and De Waal, however, argue that it was not necessary for the
court to decide on what constitutes a “fundamental precept” of the respondent’s
religion. If the respondent had waived its right to practice its religion in
this way, it would have made the decision itself. But it is in any event
doubtful that the waiver would have been binding since it cannot have qualified
as having been given in full knowledge of the freedom that is being
surrendered. In Northpine the undertaking was made in 1986, at a time
when there was no constitutionally protected right to religious freedom. It is
therefore not feasible to argue that a properly informed waiver of rights took
place, since the right in question did not exist at the time.
Direct horizontal and vertical application of the Bill of
Rights
Traditionally, a bill of rights
confines itself to regulating the “vertical” relationship between the
individual and the state. This is not a relationship of equality. The
state is far more powerful than any individual. If not protected by a bill
of rights against abuse of the state’s powers, the individual would be “in an
extremely vulnerable position.” The 1996 Bill of Rights performs this
traditional task of protecting individuals against the state by imposing a duty
on all branches of the state to respect its provisions.
The 1996 Bill of Rights goes
further than is traditional, however. It recognises that “private abuse of
human rights may be as pernicious as violations perpetrated by the state.” For
this reason, the Bill of Rights is not confined to protecting individuals
against the state. In certain circumstances, the Bill of Rights directly
protects individuals against abuses of their rights by other individuals, by
providing for the direct horizontal application of the Bill of Rights.
The direct application of the
duties under the Bill of Rights is governed by section 8. Broadly speaking,
section 8(1) deals with direct vertical application. It describes the
circumstances in which law and conduct of the state may be challenged for being
inconsistent with the Bill of Rights. Section 8(2), on the other hand, deals
with direct horizontal application. It sets out the circumstances in which the
conduct of private individuals may be attacked for infringing the Bill of
Rights. Section 8(3) grants powers to the courts to remedy such infringements.
We are concerned at this point
with direct application. Recall, though, that the Bill of Rights also applies
indirectly on both the vertical and horizontal axes. Indirect application means
that, instead of the Bill of Rights’ directly imposing duties and conferring
rights, rights and duties are instead imposed by the common law or legislation.
In turn, the development and interpretation of the common law and legislation
is influenced by the Bill of Rights.
Direct vertical application: duties of state actors
Section 8(1) provides that the
legislature, the executive, the judiciary and all organs of state are bound by
the Bill of Rights. An applicant may therefore challenge the conduct of any of
these state institutions as a breach of their duties under the Bill of Rights.
Legislatures
The term “legislature” refers to
the institutions that exercise the legislative authority of the Republic:
Parliament, the provincial legislatures and the municipal councils. The primary
duty of all of these bodies, and their principal form of conduct, is
legislating. The output of the legislative process—legislation of the central,
provincial and local governments, as well as any form of delegated
legislation—must comply with the Bill of Rights. This is because, in the words
of section 8(1), the Bill of Rights “applies to all law.”
As far as conduct of the
legislatures other than law-making is concerned, the implication of section
8(1) is that legislatures and their committees and functionaries are bound by
the Bill of Rights when they perform non-legislative functions, such as the
determination of internal arrangements, proceedings, rules and procedures. In
De Lille v Speaker of the National Assembly, the High Court stated:
The National Assembly is subject
to the supremacy of the Constitution. It is an organ of state and therefore it
is bound by the Bill of Rights. All its decisions and acts are subject to the
Constitution and the Bill of Rights. Parliament can no longer claim supreme
power subject to limitations imposed by the Constitution. It subject in all
respects to the provisions of our Constitution [... T]he nature and exercise of
parliamentary privilege must be consonant with the Constitution. The exercise
of parliamentary privilege which is clearly a constitutional power is not
immune from judicial review. If a parliamentary privilege is exercised in
breach of a constitutional provision, redress may be sought by an aggrieved
party from law courts whose primary function is to protect rights of
individuals.
Executive
The Bill of Rights binds the
“executive and all organs of state.” This means that conduct of the executive
and organs of state can be tested against any of the provisions of the Bill of
Rights, with the exception of section 33, which can only be applied to conduct
of the executive and organs of state that amounts to “administrative action.”
Although the executive and organs of state are primarily responsible for
executing the law, it must be kept in mind that the Bill of Rights also binds
these actors when they make law. All delegated legislation may therefore be
directly tested against the Bill of Rights for this reason, and for the reason
that the Bill of Rights applies to “all law.”
The “executive” may be taken to
refer to the party-political appointees who collectively head the government,
whether at the national or provincial level. At the national level of
government, for example, the executive consists of the President, the Deputy
President, the Ministers and the Deputy Ministers. On this definition, “it is
difficult to envisage conduct of the ‘executive’ that would not also amount to
conduct of an ‘organ of state’ as defined in s 239.”
Organs of state
The phrase “organ of state” is
defined in section 239 of the Constitution. In terms of this definition, the
conduct of organs of state may be divided into three categories:
- conduct of any department of
state or administration in the national, provincial or local spheres of
government;
- conduct of any other functionary
or institution exercising a power or performing a function in terms of the
Constitution or a provincial constitution; and
- conduct of any functionary or
institution exercising a public power or performing a public function in terms
of any legislation.
A court or a judicial officer is
specifically excluded from the definition.
The first category refers to any
department of state or administration in the national, provincial or local
spheres of government. When read in context, the implication of this provision
is that state departments (or the administration) are bound by the Bill of
Rights whether they exercise a power in terms of legislation or act in another
capacity. State departments will therefore be bound by the Bill of Rights when,
for example, they decide whether to enter into contracts.
By providing that the exercise of
a power or the performance of a function in terms of the Constitution, or of a
provincial constitution, amounts to conduct of an organ of state, section 239
makes it clear that the exercise of constitutional executive powers (previously
referred to as “prerogative powers”) may be challenged for consistency with the
Bill of Rights.
Finally, a functionary or an
institution qualifies as an “organ of state” in terms of s 239 when it
exercises a public power or performs a public function in terms of legislation.
This provision means, first, that the functionary or the institution must
derive powers from a statute or perform a function in terms of a statute (as
opposed to merely being incorporated pursuant to a statute, such as all
companies and close corporations are). Secondly, it means that the nature of
the power or function (and not the nature of the functionary or institution)
must be “public.” The phrase “public power” is used in section 239 of the
Constitution, but it is not defined there. It has gained wide currency in the
constitutional jurisprudence, but definition or theorisation of the concept are
seldom ventured, “because of its difficulty and abstraction.” The concept
is best understood as occupying similar terrain to the concept of “public law.”
Like public law, which operates in distinction to private law, public power
operates “in necessary but sometimes fuzzy distinction to an opposite—private
power.” Currie and De
Waal propose the following understanding of public power:
Public power is power with a
state-like dimension—either because it derives from the state or because it
does what the state typically does—exercise power in a general and
public-regarding way. The term therefore connotes use of the state’s lawfully
derived powers of regulation and compulsion. It is to be distinguished from
exercises of what can be called private power—the domain of voluntary
obligations.
Judiciary
When the members of the judiciary
(judges and magistrates) act in a judicial capacity—that is to say, when they
adjudicate legal disputes—they are required to conduct themselves in a manner
that complies with the Bill of Rights. Some provisions of the Bill of Rights,
such as section 35(5), which provides for the exclusion of evidence in certain
circumstances, are indeed specifically directed at the conduct of the judiciary
when presiding over criminal trials. When members of the judiciary perform
administrative actions, they are also bound to comply with the
administrative-justice right in section 33.
The difficult issue is to
determine the extent to which the judiciary is bound when it makes law. Every
court decision may be considered to become part of the common law and add to
the common law (unless and until it is overturned by a higher court or the
legislature). If this is so, “it can be argued that no court may give legal
effect to private conduct that is inconsistent with the Bill of Rights.” This
means that, for practical purposes, private persons will then always be bound
to the Bill of Rights, because they will be unable to seek the assistance of
the courts to enforce their unconstitutional conduct.
However, this argument has been
rejected by the Constitutional Court, on the basis that it would make section
8(2) and (3) redundant. The 1996 Constitution specifically provides that
private individuals are directly bound by the Bill of Rights in some instances,
not in every instance. This means, in effect, that common-law rules and
principles may only be directly tested against the Bill of Rights in so far as
they are relied upon by actors who are directly bound by the Bill of Rights.
Whenever such an actor, private or state, is bound, the Bill of Rights becomes
directly applicable law which overrides the common law in so far as it is
inconsistent with the Bill of Rights. In disputes between private parties
regulated by common law, the extent to which the Bill of Rights applies to
private conduct therefore determines its reach or direct application to the
common law.
Direct horizontal application: duties of private actors
The near-redundancy of direct horizontal application
Like its predecessor, the 1996
Constitution provides for direct vertical application of the Bill of Rights
but, unlike its predecessor, is not confined to this form of direct
application. Section 8(2) clearly envisages direct application of the Bill of
Rights in the horizontal relationship in certain circumstances, and therefore
“points unequivocally toward a much broader conception of direct application.”
The 1996 Constitution also still permits, however, in section 39(2) (as the
Interim Constitution did in section 35(3)), indirect application of the Bill of
Rights in horizontal cases.
The presence of section 39(2), as
Kentridge AJ stated, “prophetically,
in Du Plessis v
De Klerk, “makes much of the vertical-horizontal debate irrelevant.” Since Du
Plessis, the courts have routinely approached the issue of the effect of the
Bill of Rights on the common law indirectly. The invitation of section 8(2)—to
apply rights directly in horizontal situations—was “snubbed.”
For a while, therefore, direct
horizontality, “this deliberate innovation in the Constitution,” threatened “to
become a dead letter.” As Iain Currie and Johan de Waal observe, “Certainly,
one attraction of indirect application was that courts did not have to confront
the opacity and apparent circularity of s 8 (the Bill of Rights was to be applied
to private actors ‘where applicable’).” Whatever the reasons, indirect
horizontality provided the default form of application by which the courts
approached the common law. The trouble with this was that, besides rendering
section 8(2) of the Constitution “irrelevant,” the “model of indirect
application or, if you will indirect horizontality,” as Kentridge AJ pointed
out in Du Plessis, “seems peculiarly appropriate to a judicial system which, as
in Germany, separates constitutional jurisdiction from ordinary jurisdiction.” But,
under the 1996 Constitution, and in a deliberate alteration of the position
under the interim Constitution, South Africa no longer separates constitutional
jurisdiction from ordinary jurisdiction. Moreover, indirect application
suggests that there is a body of common law that is “conceptually separate from
the Constitution, exercising a mediating influence between the actors to whom
it applies and the Constitution. This,” write Currie and De Waal, “is difficult
to accommodate” in the remodelled constitutional system in which there is
“only one system of law.”
The question of direct
application was definitively settled by O’Regan J in Khumalo v Holomisa, “an
extremely significant decision,” where she held that “the right to freedom
of expression is of direct horizontal application” to the law of defamation. By
implication and in principle, that holding extends to other areas of private
law. This case, the Constitutional Court’s first use of the
direct-horizontality provisions of the 1996 Constitution, might be read as bringing to end
the long reign of indirect application of the Bill of Rights to the common law.
It holds (although admittedly not in so many words) that the Bill of Rights
must be applied directly to the common law wherever appropriate. It should be
directly applied, in other words, in many (perhaps most) of the horizontal
cases that have previously been treated as indirect application cases (ie,
cases involving private litigants relying on common-law provisions).
Khumalo, writes Stu Woolman,
“committed the Constitutional Court to the proposition that common-law
rules—whether challenged in disputes between the state and private parties or
in disputes between private parties—were subject to the direct application of
the Bill of Rights.”
However, despite Khumalo,
indirect horizontal application “has proven to be extremely robust and remains
the preferred judicial method for dealing with rights claims in the horizontal
dimension.”
In its only other encounter with
direct horizontality, in Barkhuizen v Napier, the Constitutional Court declined
to apply the Bill of Rights directly to a challenge to a time-limitation clause
in an insurance contract. An insurance company had rejected an insurance claim
on the grounds that, at the time of the accident, the vehicle was being used
for business purposes, despite its being insured for private use only. Two
years after the rejection of the claim, the insured issued summons against the
insurance company for the insured amount. The summons was met with a special
plea that a term of the insurance policy required any summons to be served
within ninety days of the rejection of the claim. In his replication, the
insured argued that the term requiring him to issue a summons within ninety
days was a breach of section 34 of the Constitution.
This gave the Constitutional
Court the opportunity to consider what it termed the “proper approach” to the
determination of constitutional validity of contractual clauses concluded
between private parties. The High Court had considered the matter as an
instance of direct application of section 34 to the contract, and had held that
the impugned term of the contract was in conflict with the right. The clause,
it held, was a law of general application, because it was underpinned by the
principle of pacta sunt servanda. This analysis permitted the High Court to
consider whether the limitation of section 34 by the contract was a justifiable
limitation of the right. It was held not to be; the clause was declared
invalid.
The Constitutional Court
expressed “grave doubt” about this approach, which entailed “testing the
validity of a contractual term directly against a provision in the Bill of
Rights.” Instead, the approach ordinarily to be adopted entailed indirect
application via the principle that contracts that are contrary to public policy
are unenforceable. This principle must be understood to be “deeply rooted in
our Constitution and the values that underlie it.” This meant that what public policy is and whether
a term in a contract is contrary to public policy is now to be determined by
reference to the values that underlie our constitutional democracy as given
expression by the provisions of the Bill of Rights.
Thus a term in a contract
that is inimical to the values enshrined in our Constitution is contrary to
public policy and therefore unenforceable. This approach leaves space for
the doctrine of pacta sunt servanda to
operate, but at the same time allows courts to decline to enforce contractual
terms that are in conflict with the constitutional values even though the
parties may have consented to them.
The view of Currie and De Waal is
that Barkhuizen “largely renders s 8(2) nugatory.” The Constitution applies
to all law and, in the case of the common law (“the courts’ own law”), the
default approach of the courts is to assess its constitutionality and to
develop it where necessary by way of the indirect application methodology set
out below. “The only remaining reason” to deploy direct horizontal
application would be to take advantage of the holding of the Supreme Court of
Appeal in Afrox Healthcare v Strydom, to the effect that courts in
direct-application cases are not bound by pre-1994 decisions.
See also in this regard
Barkhuizen v Napier (See its consideration especially of pacta sunt servanda) and Fraser v ABSA.
How to interpret s 8(2)
The Bill of Rights binds private
persons in certain circumstances. According to section 8(2), a provision of the
Bill of Rights applies to the conduct of a private person or a juristic person
only to the extent that the provision is applicable, taking into account the
nature of the right and the nature of any duty imposed by the right. It binds a
private or juristic person, in other words, if it is applicable to a private or
juristic person. This, it has been noted, is almost tautological.
In Khumalo, the
Constitutional Court had regard to what it described as the “intensity of the
constitutional right in question.” The meaning of this phrase, which Currie and
De Waal regard as “opaque,” appears in context “to have something to do with
the scope of the right.” The applicants were members of the media (who are
expressly identified as bearers of the constitutional right to freedom of
expression). The second factor considered by the court was the “potential of
invasion of that right by persons other than the State.” The result was a
holding that the right to freedom of expression was horizontally applicable in
a defamation case.
In the view of Currie and De
Waal, these two factors form part of a broader inquiry, consisting of five
general considerations that must be kept in mind when interpreting s 8(2):
Section 8(2) states that a
“provision” may apply to private conduct. It does not say that a “right” may
apply to private conduct. Currie and De Waal argue that “it is therefore
possible, and quite reasonable,” that some
provisions of the Bill of Rights may apply to the conduct of a private person
or juristic persons, while other provisions in the same section (and pertaining
to the same right) will not apply to such conduct. For example, the right of
access to health care services “probably does
not apply directly horizontally.” However, the right not to be refused
emergency medical treatment (s 27(3)) probably does apply horizontally. Also,
the freedom to make political choices and the right to
vote may be violated
by private conduct, but the right to free, fair and regular elections only
places duties on the state.
Questions concerning the
horizontal application of the Bill of Rights cannot be determined a priori and
in the abstract. Although this is not explicitly stated, whether a provision of
the Bill of Rights applies horizontally also depends on the nature of the private
conduct in question and the circumstances of a particular case. This explains
why section 8(2) states that a provision in the Bill of Rights binds a natural
or juristic person if, and to the extent that, it is applicable. The extent to
which a provision is applicable can only be determined by reference to the
context within which it is sought to be relied upon. For example, the right of
every arrested person to be informed promptly of the right to remain silent is
of a nature that makes it generally inapplicable to private arrests. But there
may be circumstances in which the right should apply to private arrests. There
is no reason why a private security officer, who knows of the existence of the
section-35(1)(a) right, or who may reasonably be expected to know of the right,
should not observe it. Conversely, the right to assemble peacefully and unarmed
generally applies on the horizontal level. The right to assemble in, for
example, shopping malls and on the property of an employer is therefore guaranteed.
In some circumstances, however, it may be inappropriate to apply the right
horizontally. For example, “it is unlikely that the right to assemble can be
relied on to justify demonstrations in or in front of someone’s private home.” However,
a resort to context or the circumstances of a particular case “should not be
used to frustrate the clear intention of the drafters of the 1996 Constitution” to
extend the direct operation of the provisions of the Bill of Rights to private
conduct: “It is not permissible to argue, for example, that it is only when
private persons find themselves in a position comparable to the powerful state,
that s 8(2) binds them to the Bill of Rights.” It may be that most private
or juristic persons do not have the capacity to infringe human rights in a
manner and on a scale comparable to the state, “but any interpretation of s
8(2) must avoid relying on such generalisations. The subsection was after all
included to overcome the conventional assumption that human rights need only be
protected in vertical relationships.”
The purpose of a provision is an
important consideration in determining whether it is applicable to private
conduct or not. For example, the purpose of the right to leave the Republic is,
in principle, to prevent the state from keeping persons captive in their own
country. The right to reside anywhere in the Republic is aimed at preventing
the state from reintroducing group areas-style legislation that divides the
country into racial zones. “It follows,” write Currie and De Waal, “that these
rights are not intended to have general horizontal application. On the other
hand, the purpose of the right to human dignity does not necessarily demand
differentiation between state and private conduct.” The right is to
protect an individual against assault on his or her dignity from any source,
whether private or public. The proper interpretation of a right in terms of its
purpose may therefore sometimes result in a right’s not being applicable to
private conduct, either generally or in a particular situation.
The nature of any duty imposed by
the right must be taken into account. This recognises that private or juristic
persons are often primarily driven by a concern for themselves. On the other
hand, the state is supposed to be motivated by a concern for the well-being of
society as a whole: “The application of the Bill of Rights to private conduct
should not undermine private autonomy to the same extent that it places
restrictions on the sovereignty of the government.” This consideration is
of particular importance when it comes to the imposition of duties which entail
the spending of money. Since the conduct of private persons has to be funded
from their own pockets, the same duties may not be imposed on them as may be
imposed on an organ of state which relies on public funds. For example, a
private hospital cannot (unlike a state hospital) be saddled with the duty to
provide every child with basic health care services.
In some instances, indications
are found in the Bill of Rights itself as to whether a particular right may be
applied to private conduct or not. Section 9(4), for example, states that “no
person” may discriminate, directly or indirectly, against anyone on one or more
of the grounds listed in section 9(3). Similarly, section 12(1)(c) is
explicitly made applicable to the conduct of private and juristic persons. The
section states that the right to freedom and security of the person includes
the right “to be free from all forms of violence from either public or private
sources.”
Subject to the five
considerations discussed above, “it may be said,” write Currie and De Waal, that
the nature of citizenship rights, the right to just administrative action,
and the rights of detained, arrested and accused persons, generally
preclude them from being directly applied to private conduct. Also, the nature
of the positive duties imposed by the right to have legislative and other
measures taken to protect the environment, to realise the right to
housing, the right to health care, food, water and social security, and
the right to education, “would usually result in them not being applicable
to private conduct.” The remaining rights in the Bill of Rights can,
depending on the circumstances of a particular case, be applied directly
horizontally, so as to impose duties on private individuals to conform their
conduct to the Bill of Rights.
Temporal application of the Bill of Rights
Which Constitution applies?
An unconstitutional law becomes
invalid at the moment the Constitution comes into effect. This is the effect of
the supremacy clause of the Constitution: All law and conduct inconsistent with
the Constitution is invalidated by it. When making an order of invalidity, a
court simply declares invalid what has already been made invalid by the
Constitution. This means that an unconstitutional law in force at the time of
commencement of the interim Constitution is invalidated by the interim
Constitution with effect from 27 April 1994: “If the law is challenged in
litigation brought during the period of operation of the 1996 Constitution, the
invalidity of the law should be assessed in terms of the interim Constitution.”
The doctrine described above is
known as “objective constitutional invalidity.” It means that an applicant will
always have a choice between the interim and 1996 Constitutions when
challenging old-order (pre-1994) laws. In other words, “nothing prevents an applicant
whose cause of action arose after the commencement of the 1996 Constitution
came into force from arguing that an old-order law was invalidated by the
interim Constitution.” For example, in Prince v President, Cape Law
Society, the Constitutional Court held, in litigation brought under the 1996
Constitution, that the requirement in the Supreme Court Act that eleven
judges of appeal sit in cases in which the validity of an Act of Parliament was
in question was inconsistent with the interim Constitution. According to the
Constitutional Court, the quorum requirement in the Supreme Court Act was in
conflict with the interim Constitution, which expressly provided that the
Appellate Division lacked jurisdiction to enquire into the constitutional
validity of legislation. To the extent that the Supreme Court Act provided that
the Appellate Division had jurisdiction to adjudicate the constitutionality of
Acts of Parliament, it was invalid. Moreover, it had been invalid since the
moment of commencement of the interim Constitution on 27 April 1994.
“Clearly,” write Currie and De
Waal, “there is no difficulty with the application of the rule
in Prince if the interim Constitution and 1996 Constitution contain
substantively identical provisions.” If law in force at the time of
commencement of the interim Constitution violates that Constitution, it is
invalid with effect from April 27, 1994, and will remain an invalid violation
of the 1996 Constitution, notwithstanding the repeal of the interim
Constitution by its successor. Prince, however, confronts the situation of
a law invalidated by a provision of the interim Constitution that has no
equivalent in its successor. The 1996 Constitution granted the SCA the
constitutional jurisdiction that it had been denied under the interim
Constitution, including jurisdiction to adjudicate on the constitutional
validity of Acts of Parliament. Did this mean that section 12(1)(b) was
resuscitated?
The Constitutional Court’s answer was no:
Once section 12(1)(b) became
invalid because of its inconsistency with the interim Constitution, it could
not be validated simply by the fact that under the Constitution the SCA now has
constitutional jurisdiction. Section 168(2) of the Constitution which
stipulates that the quorum of the SCA shall be determined by an Act of
Parliament must therefore, in the absence of the proviso in section 12(l)(b),
refer, at present, to section 12(1) of the Supreme Court Act which determines that
the ordinary quorum of that Court shall be five judges. This result is
consistent with the new constitutional order. Section 12(l)(b) of the Supreme
Court Act was enacted at a time when the SCA was the highest court of appeal.
That is no longer the case. Its decisions on the constitutionality of an Act of
Parliament or conduct of the President have no force or effect unless confirmed
by this Court. Its powers in this regard are therefore no different from those
conferred upon the High Court.
Currie and De Waal criticise this
explanation as “evasive” and “far from clear.” Despite this, “it is
probably safe to say that, as a general rule, a law invalidated by the interim
Constitution remains invalid after its repeal, notwithstanding any substantive
difference that there might be in the provisions of the two Constitutions.”
This, they argue, is “the logical implication” of item 2 of Schedule 6 of the 1996
Constitution: “All law that was in force when the new Constitution took effect,
continues in force.” Repeal of the interim Constitution does not deprive it of
the legal effect that it had while it was in force. One effect was the
automatic invalidation of all inconsistent law. Such law is therefore not in
force at the time of the transition to the 1996 Constitution and cannot be
resuscitated by it.
“Where the interim Constitution
is more protective than the final,” observe Currie and De Waal, “the implications
of Prince may be significant.” For example, an applicant may
choose to attack an old-order law for inconsistency with the right to freedom
of economic activity rather than relying on the narrower right to
professional freedom.
“Logically speaking,” the
doctrine of objective invalidity means that in the case of old-order
legislation, invalidity must first be assessed in terms of the interim
Constitution, notwithstanding that the cause of action may have arisen during
the operation of the 1996 Constitution. “This, however, does not happen in
practice.”
In
Ex parte Women’s Legal Centre: In re Moise v Greater Germiston Transitional
Local Council (Moise II), the court dealt with an application to amend the
order that it had made earlier, in Moise v Greater Germiston Transitional Local
Council (Moise I). In Moise I, the court had confirmed the declaration of
invalidity by a High Court of section 2(1)(a) of the Limitation of Legal
Proceedings (Provincial and Local Authorities) Act. The point made by the
applicants in Moise II was that the Limitation Act was pre-constitutional
legislation. It was found by the High Court to be a violation of the right of
access to court in s 34 of the 1996 Constitution. Section 22 of the interim
Constitution also contained a right of access to court in all relevant respects
identical to the one in s 34: “One would therefore have expected, in the light
of the Prince decision, that the subsection had become invalid at the
moment of commencement of the interim Constitution.”
The Constitutional Court
dismissed the application for an amendment on the basis that the consistency of
the Limitation Act with the interim Constitution had not been raised or
canvassed in the High Court: “Arguably, and if properly raised, a court should consider
whether a law should be declared [invalid] with reference to the interim
Constitution if it survives a challenge under the 1996 Constitution.”
The rules and principles stated
above apply to the Bill of Rights and not to the other provisions of the
Constitution.
The non-retrospectivity rule
Neither the interim nor the 1996
Constitution is retrospective in its operation. A law is retrospective if it
states that, at a past date, the law shall be taken to have been that which it
was not, so as to invalidate what was previously valid or vice versa. Neither
the interim nor the 1996 Constitution reaches backward so as to invalidate
actions taken under laws valid at the time, even if those laws were contrary to
fundamental rights. The corollary also holds: The Constitution cannot
retrospectively validate actions that were unlawful in terms of pre-1994 law.
Also, the Constitution does not interfere with rights that vested before it
came into force.
The rule that the Constitution does
not apply retrospectively affects challenges to violations of human rights that
occurred before the commencement of the Constitution. Put another way, the rule
means that a litigant can only seek constitutional relief for a violation of
human rights by conduct that occurred after commencement. As we have seen, the
implication of the doctrine of objective constitutional invalidity is that on
the date of the Constitution’s commencement, laws that are inconsistent with
the Constitution cease to have legal effect. But this does not mean that acts
performed and things done under such (unconstitutional) laws before the
Constitution came into force are also invalid. Since the Constitution does not
operate retrospectively, they remain valid. An applicant who complains about
such actions will not be allowed to challenge the constitutionality of the
enabling laws. The constitutional validity of the enabling law becomes
irrelevant since the conduct authorised by the law remains valid.
The rule of non-retrospectivity
only limits the “reach” of the Bill of Rights. In other words, it covers only
the direct application of the Bill of Rights; it does not prevent the courts
from applying the Bill of Rights indirectly to the law when developing the
common law or interpreting a statute, even if the dispute arose before the
commencement of the Constitution. This is because the post-constitutional
development of the common law, or reading down of statutes with reference to
the Constitution, does not result in the Constitution’s working retroactively.
“It is sometimes said,” the court noted in Du Plessis v De Klerk, “that
‘judge-made law’ is retrospective in its operation.” Currie and De Waal
contend that it is “always” so. It must be added, however, that the
Constitutional Court has not explicitly decided that the rule of
non-retrospectivity does not hold for the indirect application of the Bill of
Rights.
In Du Plessis v De Kierk, the
Constitutional Court expressly “left open” the question of whether a
litigant could rely on section 35(3) of the interim Constitution in respect of
a common-law claim which arose prior to the date on which the interim
Constitution came into force. Kentridge AJ nevertheless remarked that “it may
be that a purely prospective operation of a change in the common law will be
found to be appropriate when it results from the application of a
constitutional enactment which does not itself have retrospective operation.”
However, in Gardener v Whitaker,
Kentridge AJ “seemed to condone the indirect application of the Bill of Rights
to the law of defamation” in relation to an alleged defamation that took place
before the commencement of the interim Constitution.’04 Similarly, in Key v
Attorney-General a search and seizure of documents had been completed before
the interim Constitution came into force. This meant that the statutory provisions
authorising the search and seizure could not be attacked as violations of the
Constitution. Kriegler J nevertheless stated that if the evidence obtained by
way of the search and seizure was tendered in criminal proceedings against the
applicant, he would be entitled to raise Constitution-based objections to its
admissibility. While the non-retrospectivity rule prevented the applicant in
Key from challenging the provisions of the Investigation of Serious Economic
Offences Act before or during the trial, a discretion to exclude otherwise
admissible evidence could be developed by indirectly applying the Bill of
Rights.
In Masiya v Director of Public
Prosecutions, the Constitutional Court accepted that retrospective
consequences would ordinarily follow from the development of the common law in
terms of section 39(2) of the Constitution. However, it held that, on the
exceptional facts at issue—the development of the common-law definition of rape
to include anal rape of a woman—the retrospective development would offend the
principle of legality. Legality, noted the court, included the principle of
foreseeability. The rules of criminal law had to be clear, so that individuals
know which conduct is proscribed by law. The definition was accordingly held to
apply only to conduct taking place after the judgment.
Application of the Bill of Rights to matters pending at
the date of commencement
Court proceedings that commenced
prior to the coming into effect of the interim or 1996 Constitutions, but which
had not yet been finalised when those Constitutions took effect, are governed
by item 17 of Schedule 6, which provides that court proceedings that commenced
before the coming into effect of the 1996 Constitution, but after the coming
into effect of the interim Constitution, must be decided in terms of the
interim Constitution, unless the interests of justice require otherwise.
Proceedings that commenced before the coming into effect of the interim
Constitution must be dealt with in accordance with the law in force at the
time, unless the interests of justice require otherwise.
Territorial application of the Bill of Rights
“Though it is obvious,” write
Currie and De Waal, that the Constitution applies throughout the national
territory, “it is less obvious whether it has any extraterritorial
application.” The question was considered by the Constitutional Court in Kaunda
v President of the Republic of South Africa. The case arose from an incident in
which the applicants, all South African citizens, had been arrested in Zimbabwe
on charges that they had plotted to stage a coup in Equatorial Guinea. The
applicants sought relief in the form of an order directing the South African
government to seek assurances from the governments of Zimbabwe and Equatorial
Guinea that the death penalty would not be imposed on the applicants. The basis
of the application was a contention that the applicants’ constitutional rights
to a fair trial, to dignity, life and freedom and security of the person were
being infringed in Zimbabwe, and were likely to be infringed if they were
extradited to Equatorial Guinea. The state’s duty to protect the rights of the
applicants (stemming from section 7(2)) required them to be provided with
diplomatic protection.
This argument, Chaskalson CJ held
for the majority of the court, required acceptance of the proposition that “the
rights nationals have under our Constitution attach to them when they are
outside of South Africa, or that the state has an obligation under section 7(2)
to ‘respect, protect, promote, and fulfil’ the rights in the Bill of Rights
which extends beyond its borders.” According to the court, to the extent that
the Constitution provides the framework for the governance of South Africa, it
is territorially bound and has no application beyond the borders of the
Republic. As for the Bill of Rights, although foreigners are entitled to
require the South African state to respect, protect and promote their rights,
they lose the benefit of that protection when they leave the national
territory. The argument of the applicant, to the effect that section 7(2)
places a more extensive obligation on the state to respect, protect and promote
the rights of South Africans when they are in foreign countries, was rejected.
The bearers of the rights in the Bill of Rights are people in South Africa. The
Bill of Rights does not have general application beyond the national territory.
Section 7(1) does not deal with
standing, but rather with the definition of the class of beneficiaries of the
rights in the Bill of Rights. It therefore does not bar a foreign litigant who
has a protectable interest in this country from seeking to protect that
interest before a South African court.
Indirect application
Indirect application means that
the Constitution and the Bill of Rights do not directly bind actors. Instead,
the influence of the Bill of Rights is mediated through other law: statutory or
common law. In principle, and where possible, a legal dispute should be decided
in terms of the existing principles or rules of ordinary law, properly
interpreted or developed with reference to the values contained in the Bill of
Rights, prior to any direct application of the Bill of Rights to the dispute.
When it comes to statutory law, the principle simply means that a court must
first attempt to interpret legislation in conformity with the Bill of Rights
(indirect application) before considering a declaration that the legislation is
in conflict with the Bill of Rights and invalid (direct application). When it
comes to the common law, the principle supports the courts’ routine practice of
developing the common law in conformity with the Bill of Rights (indirect
application) in preference to assessing whether the common law is in conflict
with the Bill of Rights (direct application).
Indirect application to legislation: the duty to
interpret legislation in conformity with the Bill of Rights
Since the Bill of Rights binds
all the original and delegated law-making actors, it is always capable of being
applied directly to legislation. But, before a court may resort to direct
application and to invalidation, it must first consider indirectly applying the
Bill of Rights to the statutory provision by interpreting it in such a way as
to conform to the Bill of Rights.
Section 39(2) places a general
duty on every court, tribunal or forum to promote the spirit, purport and
objects of the Bill of Rights when interpreting any legislation. Statutory
interpretation must positively promote the Bill of Rights and the other
provisions of the Constitution, particularly the fundamental values in section
1. “This means,” according to the Constitutional Court in Investigating
Directorate: Serious Economic Offences v Hyundai Motor Distributors (Pty) Ltd
In re: Hyundai Motor Distributors (Pty) Ltd v Smit NO, that all statutes must be
interpreted through the prism of the Bill of Rights. All law-making authority
must be exercised in accordance with the Constitution. The Constitution is
located in a history which involves a transition from a society based on division,
injustice and exclusion from the democratic process to one which respects the
dignity of all citizens, and includes all in the process of governance. As
such, the process of interpreting the Constitution must recognise the context
in which we find ourselves and the Constitution’s goal of a society based on
democratic values, social justice and fundamental human rights. This spirit of
transition and transformation characterises the constitutional enterprise as a
whole.
The court held further that “the
purport and objects of the Constitution find expression in section 1 which lays
out the fundamental values which the Constitution is designed to achieve.” In
other words, the legislature is presumed to have intended to further the values
underlying the Bill of Rights by passing legislation that is in accordance with
the Bill of Rights, unless the contrary is established. The duty of courts,
tribunals or forums to interpret in accordance with the Bill of Rights applies
even if a litigant has failed to rely on section 39(2). The duty is therefore extensive,
“requiring courts to be alert to the normative framework of the Constitution
not only when some startling new development of the common law is in issue, but
in all cases where the incremental development of [... a common-law] rule is in
issue.”
The general duty to promote the
Bill of Rights becomes particularly important when it is possible to avoid an
inconsistency between a legislative provision and the Bill of Rights by
interpreting the legislation so that it conforms to the Bill of Rights. Under
the interim Constitution, such a process of interpretation became known as
“reading down.” According to section 35(2) of the interim Constitution, where
legislation was capable of being read in two ways—as a violation of fundamental
rights or, if read more restrictively, as not violating rights—the latter
reading was to be preferred. Section 35(2) is not repeated in the 1996
Constitution, but the courts and other tribunals are still permitted, and
indeed required, to “read down” by virtue of section 39(2). In any event,
section 35(3) of the interim Constitution, which is the predecessor to s 39(2),
always encapsulated the notion of reading down without any need for it to be
expressly spelled out in the section.
In De Lange v Smuts, Ackermann J
stated that “reading down” does no more than give expression to a
sound principle of constitutional interpretation recognised by other open and
democratic societies based on human dignity, equality and freedom such as, for
example, the United States of America, Canada and Germany, whose constitutions,
like our 1996 Constitution, contain no express provision to such effect. In my
view, the same interpretative approach should be adopted under the 1996
Constitution.
“Accordingly,” wrote Langa DP in
Hyundai Motor Distributors, “judicial officers must prefer interpretations of
legislation that fall within constitutional bounds over those that do not,
provided that such an interpretation can be reasonably ascribed to the
section.”
The duty to interpret in
accordance with the Constitution applies also where two or more interpretations
of a legislative provision are possible. The court must prefer the reading of a
statute that “better” promotes the spirit, purport and objects of the Bill of
Rights, even if neither interpretation would render the provision unconstitutional.
Methodology
In Govender v Minister of Safety
and Security, the Supreme Court of Appeal set out a standard formula for
dealing with constitutional challenges to legislation. A judge, magistrate or
presiding officer of a tribunal is required
to examine the objects and
purport of the Act or the section under consideration;
to examine the ambit and meaning
of the rights protected by the Constitution;
to ascertain whether it is
reasonably possible to interpret the Act or section under consideration in such
a manner that it conforms to the Constitution (by protecting the rights therein
protected);
if such interpretation is
possible, to give effect to it; and,
if it is not possible, to
initiate steps leading to a declaration of constitutional invalidity.
This power of interpretation,
“considerable though it is,” is not unconstrained. As the Constitutional
Court recognised in Hyundai Motor Distributors,
Limits must [...] be placed on
the application of this principle. On the one hand, it is the duty of a
judicial officer to interpret legislation in conformity with the Constitution
so far as this is reasonably possible. On the other hand, the legislature is
under a duty to pass legislation that is reasonably clear and precise, enabling
citizens and officials to understand what is expected of them. A balance will
often have to be struck as to how this tension is to be resolved when
considering the constitutionality of legislation.”
“Taken to its limit,” agree
Currie and De Waal, the power to interpret
legislation in conformity with the Constitution would mean that any legislative
provision could be made to conform to the Constitution by a suitably determined
exercise of interpretative will. This would make the powers of the courts to
declare legislation invalid superfluous and would deny the legislatures any
significant role in the interpretation of the Constitution.
Therefore, according to the
Supreme Court of Appeal, an interpretation of legislation is constrained by the
requirement that it must be “reasonably possible.” The Constitutional Court
earlier expressed the same qualification in different words in the Hyundai
Motor Distributors case: “There will be occasions when a judicial officer will
find that the legislation, though open to a meaning which would be
unconstitutional, is reasonably capable of being read “in conformity with the
Constitution”. Such an interpretation should not, however, be unduly strained.”
In National Coalition for Gay and
Lesbian Equality v Minister of Home Affairs, it was said that
there is a clear distinction
between interpreting legislation in a way which “promote[s] the spirit, purport
and objects of the Bill of Rights” as required by s 39(2) of the Constitution
and the process of reading words into or severing them from a statutory
provision which is a remedial measure under s 172(1)(b), following upon a
declaration of constitutional invalidity under s 172(1)(a) The first process,
being an interpretative one, is limited to what the text is reasonably capable
of meaning. The latter can only take place after the statutory provision in
question, notwithstanding the application of all legitimate interpretative
aids, is found to be constitutionally invalid.
“It follows,” held the court in
Hyundai Motor Distributors, “that where a legislative provision is reasonably
capable of a meaning that places it within constitutional bounds, it should be
preserved. Only if this is not possible should one resort to the remedy of
reading in or notional severance.”
These qualifications may be taken
to mean something along the lines of “plausible”—the result of the
interpretative process must be a reading of the legislation that is defensible
using “all legitimate interpretive aids,” by which is meant the repertoire of
justificatory arguments supplied by the law of interpretation of statutes. The
process of reaching a plausible, constitutionally-compliant interpretation
entails reading legislation purposively and contextually. If the only plausible
interpretation of a statutory provision is one that entails an infringement of
the Bill of Rights, the court must proceed to consider whether the infringement
is justifiable. If is not, it must declare the provision unconstitutional.
Interpretation in conformity is not confined to the
restrictive reading of legislation
“Reading down” should perhaps be
avoided as a description of the practice of interpretation in conformity with
the Constitution, as it tends to suggest that the practice always entails
reading restrictively. But section 39(2) sometimes requires more than simply
narrowing the ambit of legislation so as to avoid conflict with rights. A
narrow construction of a legislative provision will often have the result of
avoiding an alleged conflict between the provision and the Bill of Rights: for
example, when discretionary powers conferred are too wide, or when the scope of
regulation is over-inclusive, or to restrict the impact of changes to the
criminal law. But, on other occasions, the statute may have to be generously interpreted
to avoid the conflict: for example, where the constitutional invalidity lies in
the lack of any express grant of discretionary power. The point is that, if the
statutory provision is genuinely ambiguous or otherwise unclear, the
interpretation which best conforms to the Bill of Rights must be chosen.
Section 39(2) does not have any
bearing on the interpretation of the Constitution or the Bill of Rights; the
subsection deals with statutory interpretation only.
Indirect application of the Bill of Rights to disputes
governed by the common law
The obligation to develop the common law
As we have seen, legislation is
approached by first interpreting it with the Constitution in mind, prior to any
direct application of the Constitution (and any finding of
unconstitutionality). In the case of the common law, the approach is similar
but not identical, the difference lying in the remedial powers of the courts.
If impugned legislation is found to limit a right, and if the limitation does
not satisfy the justification standard in section 36, the court provides a
remedy by declaring the legislation unconstitutional and, where possible,
ameliorating the constitutional defect through reading in or notional or actual
severance. “In that event,” according to Moseneke J in S v Thebus, “the
responsibility and power to address the consequences of the declaration of
invalidity resides, not with the courts, but pre-eminently with the legislative
authority.” The Constitutional Court was unanimous on this issue. In Du Plessis
v De Klerk, furthermore, the court noted that “the common law, it is often
said, is developed on incremental lines. Certainly it has not been developed by
the process of ‘striking down.’”
But the common law is different.
It is the law of the courts and not the legislature:
The superior courts have always
had an inherent power to refashion and develop the common law in order to
reflect the changing social, moral and economic make-up of society. That power
is now constitutionally authonsed and must be exercised within the prescripts
and ethos of the Constitution.
According to the court, the need
to develop the common law under section 39(2) could arise in at least two
instances:
The first was when a rule of the
common law is inconsistent with a constitutional provision. Repugnancy of this
kind would compel an adaptation of the common law to resolve the inconsistency.
The second possibility was that
“a rule of the common law is not inconsistent with a specific constitutional
provision but may fall short of its spirit, purport and objects.” If so, “the
common law must be adapted so that it grows in harmony with the ‘objective
normative value system’ found in the Constitution.”
In a constitutional challenge of
the first type the court must perform a “threshold analysis,” being whether the
rule limits an entrenched right, if the limitation is not reasonable and
justifiable, the court itself is obliged to adapt, or develop the common law in
order to harmonise it with the constitutional norm.
In its earlier decision in
Carinichele, the Constitutional Court emphasised that the constitutional
obligation to develop the common law is not discretionary, but is rather a
“general obligation” to consider whether the common law is deficient and, if
so, to develop it to promote the objectives of the Bill of Rights. The
obligation applied in both civil and criminal cases, irrespective of whether or
not the parties had requested the court to develop the common law.
The methodology of indirect application to the common law
The indirect application of the
Bill of Rights to the common law can take many forms. The first method is to
argue for a change in the existing principles of the common law so that the law
gives better effect to Bill of Rights. This argument has been made in the areas
of defamation and restraint of trade. In restraint of cases the argument that
the incidence and content of the onus have to be reformed with reference to the
section 22 right to occupational freedom has been less successful. The courts
have been less inclined to reform the principles of the law of contract in a
similar manner to the development of the law of delict.
The second method is to “apply”
the common law with due regard to the Bill of Rights. This method was employed
by Davis AJ in Rivett-Carnac v Wiggins. Davis AJ declined to consider the
constitutionality of the presumption relating to animus iniuriandi in
defamation cases, but “clearly took the Bill of Rights into account in reaching
the conclusion that the statements made in this particular case were not
defamatory.” Davis AJ held that the “boundary between criticising
professional work without reducing such professional’s reputation in the eyes
of colleagues and the publication of defamatory statements about such a
professional must be carefully drawn, particularly in the light of our new
constitutional commitments.”
The third method, which is
closely related to the second, is to give constitutionally-informed content to
open-ended common-law concepts, such as “public policy” or “contra bonos mores”
or “unlawfulness.” This has been held to be the proper approach to Bill of
Rights challenges to contractual provisions. Barkhuizen v Napier dealt with a
contractual time-bar clause requiring action to be instituted against an
insurer within ninety days of the rejection of an insurance claim. The insured
contended that this clause infringed his right of access to court in terms of
section 34. The approach to be adopted to this contention was to apply a
“constitutionalised” conception of the common-law doctrine of public policy: a
conception informed by the values given effect to in the Bill of Rights.
The analysis then undertaken by
the Constitutional Court in Barkhuizen essentially entailed an application of
the test for the validity of legislative time-bar provisions set out in Mohlomi
v Minister of Defence: A time-bar provision will unjustifiably limit the right
of access to court if it is unreasonably short and if it is inflexible. If it
is an unjustifiable limitation of section 34, it will be contrary to public
policy and unenforceable. The justifiability of the provision had to be
determined in the light of a number of factors, including the bargaining
position of the respective parties and their ability to enforce their rights.
In Barkhuizen, the court found
that the applicant was well-resourced and there was nothing on the facts to
explain why no steps had been taken to enforce his rights at the earlier
stages. Similarly, the facts did not disclose the extent of the bargaining
between the parties prior to the conclusion of the contract. It could not be
said, for instance, whether the insured was in a weak bargaining position and
could not influence the terms of the contract.
Bredenkamp v Standard Bank is an
application of the Barkhuizen methodology. The issue was whether the
contractual right of a banker to close a client’s account was subject to the
requirements of fairness. The Supreme Court of Appeal held that the Constitution
does not envisage the duty of fairness to apply in all contractual settings.
Rather, one must consider the specific circumstances of each case to determine
whether a constitutional value is implicated. If not, one cannot complain about
an overarching requirement of fairness: “If a contract is prima facie contrary
to constitutional values, questions of enforcement would not arise. However,
enforcement of a prima facie innocent contract may implicate an identified
constitutional value. If the value is unjustifiably affected, the term will not
be enforced.”
This means that fairness is not a
self-standing requirement against which contractual clauses must be assessed:
“Fairness is part of a matrix of constitutional values, which inform the
interpretation of contracts. Such values are an embodiment of the legal
convictions of the community. The autonomy of individuals to freely conclude
contracts which are binding upon them is also part of the legal convictions of
the community.”
Limits on indirect application to the common law
A rule of the common law must be
assessed for inconsistency with the Bill of Rights and, if necessary, developed
within the “matrix of [... the] objective, normative value system” established
by the Constitution. Courts have far more scope to “develop” the common law by
way of indirect application than they have when they “interpret” legislation,
where they are bound to a reasonable interpretation of the statute.
Are there any limits on the power
to develop the common law? The first limitation is that, when the common law is
developed, it must be done incrementally and on a case by case basis. The
development cannot take place in the abstract; the court must apply the law as
it is found to be in the case before it. This approach has also found favour
when the Bill of Rights is directly applied to the common law. Indeed, it is
even more important when the Bill of Rights is directly applied, because the
consequences of a direct application differ from those of an indirect
application. For example, in Shabalala v Attorney-General, Transvaal, the
Constitutional Court was careful, after striking down a common-law rule, to
balance the need to provide guidance with the danger of being prescriptive.
Such care must also be taken when the Bill of Rights is indirectly applied.
“Some guidance on the new approach has to be provided,” write Currie and De
Waal, “while room must be left for the courts to develop the principle on a
case by case basis.”
Stare decisis and indirect application
“One of the most important” limitations
on the power to develop the common law via the indirect application of the
Constitution is the doctrine of stare decisis. In Govender v Minister of Safety
and Security, reading down was employed to hold that section 49(l)(b) of the
Criminal Procedure Act was not unconstitutional. In a subsequent decision, the
Transkei High Court, in S v Walters, confronted with the precedent of the SCA
decision in Govender, held that it did not have to follow it. Appeal-court
decisions on the constitutional validity of legislation, according to Jafta AJP,
“rank in the same level” as High Court decisions. The reason is that both
decisions had no force unless confirmed by the Constitutional Court. Since, in
the view of Jafta AJP, the SCA’s decision on section 49(1)(b) in Govender was
clearly wrong, it did not have to be followed by the High Court. The subsection
was struck down to the extent that it permitted the use of force to prevent a
suspect from fleeing.
The High Court’s approach to the
issue was repudiated by the Constitutional Court in the confirmation
proceedings:
The trial court in the instant
matter was bound by the interpretation put on section 49 by the SCA
in Govender. The judge was obliged to approach the case before him on the
basis that such interpretation was correct, however much he may personally have
had his misgivings about it. High courts are obliged to follow legal
interpretations of the SCA, whether they relate to constitutional issues or to
other issues, and remain so obliged unless and until the SCA itself decides
otherwise or [... the Constitutional Court] does so in respect of a
constitutional issue.
But this holding, Kriegler J
emphasised, applied only to the binding effect of decisions of higher tribunals
“delivered after the advent of the constitutional regime and in compliance with
the requirements of section 39 of the Constitution.” The extent of application
of stare decisis to pre-1994 decisions (if this is what “the advent of the
constitutional regime” means), and to direct applications of the Constitution,
was not decided.
The subsequent decision of the
Supreme Court of Appeal in Afrox v Strydom “fills the gap left open by the
Constitutional Court.” As regards the
binding effect of pre-constitutional authority of the appeal court, there are
three distinct situations that can arise:
Direct application of the
Constitution to the common law: “The High Court is convinced that the relevant
rule of the common law is in conflict with a provision of the Constitution.” In
such situations, pre-Constitutional authority is not binding on a High Court.
Pre-constitutional decisions of
the appeal court based on open-ended considerations such as boni mores or public interest: In such
situations, the High Court may depart from earlier authority if convinced,
taking the values of the Constitution into account, that it no longer reflects
the boni mores or the public
interest.
The third situation is that of an
indirect application of the Constitution to the common law, by way of section
39(2). Even if convinced that the rule must be developed to promote the spirit,
purport and objects of the Bill of Rights, a High Court is obliged to follow
the authority of pre-constitutional decisions of the appeal court.
Currie and De Waal put Afrox and
Walters together in the following way:
Post-constitutional decisions of
higher courts are binding, whether they are on constitutional issues or not.
Pre-1994 decisions of higher
courts on the common law are binding, except in cases of direct conflict with
the Constitution or in cases involving the development of open-ended standards
such as boni mores.
The distinction between direct
and indirect application is therefore “crucial to the impact of
the Afrox decision.” Section 39(2), the SCA holds, does not
authorise lower courts to depart from higher authority, whether pre- or
post-constitutional. The subsection must be read with section 173, recognising
the inherent jurisdiction of the High Courts to develop the common law. It is
that power which is exercised when the courts develop the common law in
accordance with section 39(2). But the power has always been constrained by the
doctrine of stare decisis: “There is nothing to indicate that the Constitution
has changed this.”
The Afrox and Walters decisions
have been strongly criticised. There is, however, “a significant omission from
the Afrox decision.” As we have seen, indirect application in
terms of section 39(2) does not involve only development of the common law, but
also statutory interpretation, taking the spirit, purport and objects of the
Bill of Rights into account. But the SCA in Afrox “seems to confine itself to
the first type of indirect application only.” This may be taken to mean
that “post-Afrox High Courts still possess the jurisdiction to depart from
pre-constitutional statutory interpretations of the AD.”rrie and De
Waal argue that “a great deal also turns on the distinction between direct and
indirect application.” Khumalo v Holomisa appears to treat direct
horizontal application as a relatively simple and unexceptional process. If so,
“awkward appeal court precedent can easily be sidestepped. A High Court, by
opting for direct application, will be understood to distinguish the case
before it from a precedent arising from indirect application.”
Manner of application of the Bill of Rights in legal
disputes
Under the 1996 Constitution,
there is only one system of law. The Constitution applies to all law, informing
its interpretation and development by the courts and determining its validity.
This means that the parallel systems of “constitutional” law and
“non-constitutional” law (and “constitutional” and “non-constitutional litigation”)
developed under the interim Constitution are no longer theoretically
sustainable. Nevertheless, the distinction between the direct and indirect
methods of application of the Constitution to the law has not been abandoned
and “continues to have some practical significance at least in so far as the
common law is concerned.”
Jurisdiction
We have seen that, under the
interim Constitution, the distinction between direct and indirect application
of the Bill of Rights had important jurisdictional implications. The interim
Constitution distinguished between constitutional matters and other matters,
and provided that the Constitutional Court could hear only the former and the
Appellate Division only the latter. In Du Plessis, the Constitutional Court
held that indirect application of the Bill of Rights to the common law was not
a constitutional matter, and therefore was within Appellate Division
jurisdiction. The main task of the Constitutional Court was to test the
validity of the law and state conduct against the Constitution. In order to
trigger the jurisdiction of the Constitutional Court, it was therefore
necessary to show that the Bill of Rights applied directly to the challenged
law or conduct. Whenever the Bill of Rights merely applied indirectly to a
dispute, the Appellate Division and not the Constitutional Court was primarily
responsible.
Under the unitary jurisdictional
system established by the 1996 Constitution, all superior courts have the power
to apply the Constitution directly and indirectly to the common law. This means
that the jurisdictional motivation for distinguishing between direct and
indirect application no longer holds for common-law disputes. However, since
decisions of the High Courts and the Supreme Court of Appeal declaring certain
forms of legislation invalid must be confirmed by the Constitutional Court, “it
remains important for jurisdictional reasons,” whether legislation is
directly tested against the Bill of Rights, or whether it is merely interpreted
with reference to the Bill of Rights.
The purpose and effect of direct application differ from
that of indirect application
The purpose of direct application
is to determine whether there is, on a proper interpretation of the law and the
Bill of Rights, any inconsistency between the two. The purpose of indirect
application is to determine whether it is possible to avoid, in the first
place, any inconsistency between the law and the Bill of Rights by a proper
interpretation of the two.
Direct application of the Bill of
Rights generates a constitutional remedy, whereas indirect application does
not. The reason for this is that direct application is aimed at exposing
inconsistency between the Bill of Rights and law or conduct. If there is, the
court then declares that law or conduct constitutionally invalid. The effect of
such a declaration, according to Ackermann J and Sachs J in Du Plessis, is to
restrict the legislature’s options in amending the law or enacting a similar
law. Much depends of course on the terms of the court’s order and its reasoning
and the application of the doctrine of stare decisis, but as a general rule
direct application rules out certain possibilities as constitutionally
impermissible, whereas an indirect application merely proposes a construction
of the law that conforms to the Constitution. Although there is, therefore, a
difference in principle between direct and indirect application, the problem
alluded to by Ackermann J and Sachs J also depends on the extent to which a
court is prepared to “pronounce on the meaning” of the Constitution: “Courts
generally avoid making extensive pronouncements on what the Constitution
demands the common law to be, whether they apply the Bill of Rights directly or
indirectly.” The preferred approach is to give narrow rulings limited to
the facts before the court: “Such orders will preserve considerable space for
the legislature to reform the common law.” Direct application, however,
“inevitably rules out certain options.” When a law or conduct is ruled to
be inconsistent with the Constitution, it can no longer form part of the law.
The scope of the limitation on the legislature’s discretion will therefore
depend on the extent of the court’s ruling.
That said, there is little
practical difference between the two forms of application when it comes to the
common law. This is because, although notionally methodologically distinct,
direct and indirect application of the Bill of Rights end up at the same point:
the need to develop rules of the common law in conformity with the Bill of
Rights.
“There are,” observe Currie and
De Waal, “only a few common-law cases where the method of application is likely
to make a substantive difference to the result.” These are cases in which
a plaintiff cannot find a cause of action in the existing common law. Since the
common law does not provide a right, it will be necessary to invoke directly a
right in the Bill of Rights.
Indirect application must be considered before direct
application
In S v Mhlungu, Kentridge AJ
stated, I would lay it down as a general
principle that, where it is possible to decide any case, civil or criminal,
without reaching a constitutional issue, that is the course which should be
followed.
This statement was subsequently
approved by the unanimous court in Zantsi v Council of State, Ciskei. In this
case, Chaskalson P referred to the “salutary rule” which is followed in the
United States “never to anticipate a question of constitutional law in advance
of the necessity of deciding it” and “never to formulate a rule of
constitutional law broader than is required by the precise facts to which it is
to be applied.” This rule, Chaskalson P added,
allows the law to develop
incrementally. In view of the far-reaching implications attaching to
constitutional decisions, it is a rule which should ordinarily be adhered to by
this and all other South African Courts before whom constitutional issues are
raised [.... I]t is not ordinarily desirable for a Court to give rulings in the
abstract on issues which are not the subject of controversy and are only of
academic interest.
There are several reasons for
observing this “salutary rule” under the South African Constitution. The first
is procedural. The interim Constitution contained complicated provisions
governing the referral of a constitutional issue to the Constitutional Court
where that issue was beyond the jurisdiction of the Supreme Court. A referral
was, for example, necessary whenever the constitutionality of an Act of
Parliament was in dispute. The statements in Mhlungu and Zantsi “were meant to
deter the divisions of the Supreme Court from referring irrelevant issues or
issues that were within their jurisdiction to the Constitutional Court.” Since
the system of referrals has now been replaced by a wider High Court
jurisdiction and a system of appeals, this justification should no longer carry
the same weight. However, it would be wrong to conclude that the justification
did not survive the changes in constitutional jurisdiction brought about by the
1996 Constitution and the abolition of referrals. It remains an important
factor when considering applications for direct access to the Constitutional
Court and applications for leave to appeal using the “leapfrog” appeal
procedure. It also informs the doctrine of justiciability, particularly the
principles that courts should not decide moot cases or cases that are not ripe
for judicial resolution.
There are also substantive
reasons for observing the rule: “Courts should avoid making pronouncements on
the meaning of the Constitution where it is not necessary to do so, so as to
leave space for the legislature to reform the law in accordance with its own
interpretation of the Constitution.” Lengthy expositions of the
Constitution may result in actual or perceived restrictions on the legislature,
a “constitutional straitjacket” which makes it difficult for the legislature to
respond to changing circumstances. The courts, and particularly the
Constitutional Court, are not the only interpreters of the Constitution. They
are, however, its final and authoritative interpreters. Before pronouncing on
the meaning of the Constitution, “the courts should allow other organs of the
government the opportunity to interpret and give effect to the Constitution.” Practically,
this means that the legislature should be given the opportunity to address an
issue before a court decides on it: “The legislature and the executive are
better equipped to ascertain the needs of society and to respond to those
needs.” Once such a response finds expression in legislation, courts may
then test the legislation against the provisions of the Bill of Rights. Even
then, the Constitutional Court (the final court in constitutional matters)
often seeks to avoid ruling on the constitutionality of a statutory provision
until trial, and appeal-court judges have expressed their views on the effect
of the provision and the likely consequences of invalidating it. “It is sound
judicial policy,” write Currie and De Waal, “to decide only that which is
demanded by the facts of a case and that is necessary for its proper
disposition; this allows constitutional jurisprudence to develop
incrementally.”
When applying the Bill of Rights
in a legal dispute, the principle of avoidance is “of crucial importance.” As
we have seen, the Bill of Rights always applies in a legal dispute. It is
usually capable of direct or indirect application and, in a limited number of
cases, of indirect application only. The availability of direct application is
qualified by the principle that the Bill of Rights should not be applied
directly in a legal dispute unless it is necessary to do so. The principle has
“a number of important consequences.”
Even when the Bill of Rights
applies directly, a court must apply the provisions of ordinary law to resolve
the dispute, especially in so far as the ordinary law is intended to give
effect to the rights contained in the Bill of Rights. Many recent statutes,
such as the Labour Relations Act 66 of 1995 and the Equality and Administrative
Justice Acts are intended to implement the Bill of Rights. They must first be
applied, and if necessary interpreted generously to give effect to the Bill of
Rights, before a direct application is considered.
The same applies to disputes
governed by the common law. The ordinary principles of common law must first be
applied, and if necessary developed with reference to the Bill of Rights,
before a direct application is considered.
When the Bill of Rights is
directly applied in disputes governed by legislation, conduct must be
challenged before law. In other words, the implementation of the statute must
be challenged before the provisions of the statute itself.
However, “to complicate matters
further,” the principle that constitutional issues should be avoided
is not an absolute rule. It does not require that litigants may only directly
invoke the Constitution as a last resort. As with many legal principles, its
force depends on the circumstances of the case. Where the violation of the
Constitution is clear and directly relevant to the matter, and there is no
apparent alternative form of ordinary relief, it is not necessary to waste time
and effort by seeking a non-constitutional way of resolving a dispute. This
will often be the case when the constitutionality of a statutory provision is
placed in dispute because, apart from a reading down, there are no other remedies
available to a litigant affected by the provision. On the other hand, the
principle of avoiding constitutional issues is particularly relevant when the
interest of an applicant in the resolution of a constitutional issue is not
clear, and where the issue is not ripe for decision, or when it has become
academic or moot.
