Issued by the FW de Klerk Foundation 15/03/2024 


Understanding the different versions:

The version of the General Intelligence Laws Amendment Bill (“the GILAB”) that was opened for public comment, was first introduced into Parliament by the Minister in the Presidency on 17 November 2023. Parliament’s first house, the National Assembly (NA), then asked the public for their input from 17 December 2023 to 15 February 2024.

An earlier draft version of the Bill was approved by Cabinet on 24 May 2023, for submission to Parliament. This version, which explicitly mandated security vetting for religious organisations and non-governmental organisations (“NGOs”), was never introduced or allocated to the Parliamentary committee working on the GILAB, nor opened for public comment.

The version of the Bill that was, in fact, tagged, allocated to the committee and opened for public comment, replaced religious organisations and NGOs with the now infamous phrase: “person or institution of national security interest”.

What the (current version of the) GILAB aims to do:

The GILAB aims to, amongst other things, split the State Security Agency into two distinct arms: the SA Intelligence Agency (which will have a domestic focus) and the SA Intelligence Service (which will be responsible for foreign intelligence). (This splitting is in line with the recommendations of the High-level Review Panel on State Security Agency.)

The GILAB also seeks to be the law that regulates “bulk surveillance”, after the Constitutional Court in Amabhungane Centre v Minister of Justice found the practice to be illegal, because there was no law authorising it.

What are the problems with the (current version of the) GILAB?

While GILAB arguably does aim to do some good (e.g. splitting the foreign and domestic branches of State Security Agency), it is all the other things it wishes to do that have been the cause for such grave concern across society – from human rights organisations, journalists, religious institutions, private businesses, etc. In short, these problems are as follows:

  1. Problematic definitions;
  2. Mandatory vetting; and
  3. Non-compliance with a Constitutional Court judgment.


Problem 1: GILAB’s definitions are MIA, circular, too wide and/or vague


Example: “threat” the most crucial concept underpinning all the intelligence laws the GILAB wishes to change and the GILAB itself, is nowhere defined.


Example: “national security” and “opportunity or potential opportunity” are circular definitions which rely on themselves within the same context, creating an endless loop of interpretation. Thus, they are whatever the SA Intelligence Agency deems them to be.

Too wide:

Example 1: the Bill’s definition of “threat to national security” fails to exclude and protect “lawful political activity, advocacy, protest or dissent”. Anyone who engages in these activities, which should be protected as part of one’s political rights (section 19 of the Constitution), may therefore be seen by the Intelligence Agency as a “threat to national security”.

Example 2: The factors currently contained in the Bill’s definition of “security competency test” currently read disjunctively: “administering a vetting investigation to determine the security competence of a person or institution and if such person or institution is suitable to access classified information or critical infrastructure of the State or is viewed as vulnerable to blackmail, undue influence or manipulation or security compromise or is a person or institution of national security interest…”

This disjunctive reading (i.e. use of “or” as opposed to “and”) of the factors, makes the definition of “security competency test” too wide, because literally every citizen in the Republic can be deemed to be vulnerable to e.g. “manipulation”. However, that does not mean that they are a threat to national security and that the State should be able to administer a security competency test on them. It also results in the clause failing to achieve its purpose (which is no doubt to only vet individuals with access to classified information and critical State infrastructure, not every citizen in the Republic). This failure, in turn, results in an arbitrary infringement of constitutional rights which makes it an unjustifiable and unreasonable limitation of these rights.


Example 1: The definition of “person or institution of national security interest” is extremely concerning, because it rests on two (2) problematic concepts, namely:

  1. The Bill says these are people or institutions whose activities are deemed by the Intelligence Agency to be inconsistent with South Africans’ resolve to live as equals; live in peace and harmony; be free from fear and want; and to seek a better life (i.e. contrary to section 198 of the Constitution). These vague and ambiguous concepts mean that potentially anything could be (mis)construed as contrary to section 198 and, therefore, anyone who catches the Agency’s interest could be said to be a person or institution of national security interest.
  2. These activities include, but are not limited to, activities that are defined as a “threat to national security”, which is problematic because it fails to exclude “lawful political activity, advocacy, protest or dissent”.

Example 2: the phrase “opportunity or potential opportunity or threat or potential threat to national security” is used in many of GILAB’s definitions (e.g. “Domestic Intelligence”; “Foreign Intelligence”; “intelligence gathering”; “national security”; “national security intelligence” etc.). Not only is this phrase redundant (an opportunity, after all, is a potential good, so a potential opportunity is a potential, potential good), but it is a vague concept.

The result of all these definitions is that they contravene the Constitution (either specific rights or the founding value of the rule of law) and are unworkable in practice.

Solutions to problem 1:

  1. Define “threat” as “impending danger of serious harm to the Republic as one, sovereign, democratic state founded on the values set out in section 1 of the Constitution” and delete all references to “opportunity or potential opportunity“. This will solve many problematic aspects, including, but not limited to various definitions, and aspects related to, and the mandatory vetting provisions.
  2. Tighten and clarify the overbroad and circular definitions as follows:

Problem 2: Mandatory Vetting

In addition to all these problematic definitions, the Bill makes it compulsory for the Intelligence Agency to conduct vetting investigations on persons or institutions suspected of being threats or potential threats to national security. This is done to determine their security competence for the purpose of deciding whether they qualify for a security clearance certificate, or not.

The problem arises from the absence of a clear definition for “threat” and the circular definition of “national security”. Consequently, this empowers the Intelligence Agency to conduct vetting investigations on entities, such as media houses or civil society organisations, including their employees.

Solution to problem 2:

Limit mandatory vetting to only apply to categories of applicants and employees of organs of the State and Departments of State.

Problem 3: Noncompliance with the Constitutional Court’s Amabhungane Centre v Minister of Justice judgment by failing to provide for post-surveillance notification

In this judgment, the Constitutional Court held that the Regulation of Interception of Communications and Provision of Communication-Related Information Act, 2002 (“RICA”) was unconstitutional, because it failed to provide a post-surveillance notification to the person, who had been the subject of the surveillance, that they had been surveilled. This failure amounted to an unjustifiable infringement of their constitutional rights to privacy and access to justice.

The Bill contains the same failure and is, therefore, unconstitutional.

Solution to problem 3:

Provide for post-surveillance notification.


The above three (3) problems contained in the version of the GILAB that was opened for public comment, will result in the State being able to spy on anyone without ever telling them. These problems make the Bill open to abuse by future governments wishing to threaten or even silence critical voices who oppose it.

This is why the FW de Klerk Foundation has been steadfast that if the State wishes to limit a constitutional right (e.g. privacy, political rights etc.), it may only do so following the Constitution’s section 36 limitations test, which the Bill fails. The Foundation has also consistently pointed out to Parliament that due to the circularity of many of its definitions, it also fails to comply with a founding value of the Constitution, namely the rule of law (section 1(c)), which requires rationality.