Judicial review in South African administrative law is never dull, but it is currently even more interesting than usual. One reason for this is the ongoing contest between the two main avenues to judicial review: the Promotion of Administrative Justice Act 3 of 2000 (PAJA) – constitutionally mandated if rather unpopular legislation produced by the democratic legislature – and the constitutional principle of legality, the creation of a highly engaged and percipient judiciary. The principles at stake include subsidiarity, constitutional supremacy and the separation of powers.
The contest and its outcome will be of special interest to ‘constitutionalised’ jurisdictions and those that have codified their grounds of review. But it is of more general interest too, for the development of the principle of legality is another illustration of the vigour and fecundity of the rule of law.
The PAJA was enacted to give effect to the fundamental rights to lawful, reasonable and procedurally fair administrative action contained in section 33 of the 1996 Constitution. Section 6 of the statute provides for judicial review on a wide range of grounds. But the PAJA, like section 33, applies only to ‘administrative action’, and not every exercise of public power qualifies as such in terms of the narrow and complicated statutory definition of the concept.
The Constitutional Court anticipated this problem before the PAJA came into existence, and in the Fedsure case (1998) it identified a constitutional principle of legality as a route to judicial review for all exercises of public power. It characterised the principle as an aspect of the rule of law (a founding value of the 1996 Constitution) and a more general and abstract version of the right to lawful administrative action (paras 56–59).
The legality principle was evidently envisaged as merely a residual pathway allowing for review of executive action and other kinds of non-administrative action on limited grounds, and at first it required merely that actors remain within the bounds of their powers (Fedsure para 58). But the principle burgeoned: it soon came to require more elaborate kinds of lawfulness and a minimum level of rationality. The idea of rationality expanded, too, and was held to imply the need for procedural fairness and reason-giving in appropriate cases. In short, the courts made the principle of legality mean whatever they wanted it to mean as they went about creating a sort of common law for the constitutional era.
The result today is a closer resemblance between the content of the legality principle and the PAJA grounds of review, giving litigants much less incentive to proceed by way of the PAJA. And there are plenty of disincentives. Apart from its complicated definition of administrative action, the PAJA imposes unpopular procedural rules on applicants for judicial review: an outer limit of six months for making the application (section 7(1)) and a stringent duty to exhaust internal remedies first (section 7(2)). By contrast, legality review attracts more indulgent common-law versions of these rules, which adds to its allure.
It is understandable, then, that applicants have been relying on legality more and more often at the expense of the PAJA. What is extraordinary is that the courts have largely been going along with often blatant avoidance of constitutionally mandated legislation .
While the Constitutional Court made it clear more than a decade ago that the PAJA cannot legitimately be avoided by direct reliance on section 33 or on the common law, it was unhelpfully silent about sidestepping the PAJA in favour of the principle of legality. Then, in its breathtaking judgment in the Albutt case (2010), the court held that there was no need to resort to the PAJA at all where the legality principle was capable of resolving a dispute. More than that, it criticised the court below for engaging in the administrative action inquiry, a merely ‘ancillary’ question that there was no need for the court to ‘reach’ precisely because the case could be resolved by the legality principle (paras 82–83).
This remarkable line of reasoning ignored the principle of subsidiarity in adjudication, which requires more specific and detailed norms to be applied in preference to more general and abstract ones. It disrespected the mandate in s 33(3) of the Constitution and the legislature’s response to that mandate. Indeed, the court’s reasoning would ineluctably have led to the redundancy of the PAJA.
In Minister of Defence v Motau (2014) the Constitutional Court admitted that the PAJA must be applied in preference to the legality principle. However, it buried this important point in a footnote (para 27 fn 28) and, disappointingly, failed to mention the contrary approach in Albutt. Frankly, some of us wondered whether litigants or other courts would take any notice – and that was one reason to welcome the majority judgment of the Supreme Court of Appeal (SCA) in State Information Technology Agency v Gijima Ltd.
The applicant, SITA, asked the court to set aside as unlawful SITA’s own decision to enter into a contract with Gijima pursuant to a procurement process. Being seriously out of time, SITA ignored the PAJA and its inconvenient time limit and relied on the legality principle instead. The court below held that it was not entitled to avoid the PAJA, and a majority of the SCA agreed – loudly and clearly. Cachalia JA said (paras 38 and 44):
In my view, the proper place for the principle of legality is to act as a safety-net or measure of last resort . . .. [W]hen PAJA does apply, litigants and the courts are not entitled to bypass its provisions and rely directly on the principle of legality.
The majority rejected SITA’s argument that as an organ of state it was not bound by the PAJA and/or by section 7(1), and was thus free to use the legality principle. The minority, on the other hand, seemed to accept this argument, and deplored the majority approach as ‘slavish adherence to formalism’ and as ‘compromising substance’ (para 55). But the minority may not have been entirely convinced of its own position: it veered from asserting confidently that the use of legality was ‘the proper route to take in this case’ (para 68) to the weaker claim that it would be ‘in the public interest to allow Sita to vindicate . . . the principle of legality and not to thwart it by procedural technicalities’ (para 70). And that claim had a hollow ring in any event, for on the majority view SITA’s true reason for going to court was not to vindicate legality but the dishonourable and self-interested one of avoiding arbitration of a dispute about payment (para 39). The majority aptly quoted Boonzaier’s observation that officials are capable of acting ‘antithetically to the rule of law even as they purport to assert legality’.
A noteworthy feature of Gijima is the absence of any reference to Albutt and Motau. There is no mention of the Constitutional Court’s actively encouraging avoidance of the PAJA in Albutt, and no allusion to the Motau footnote that points precisely in the opposite direction. While neither of those cases involved an application by an organ of state to review its own decision, both are undeniably pertinent to the relationship between PAJA and the legality principle. Some of us are daring to hope that the highest court will confront those cases when it hears the Gijima appeal in May 2017, and that the court will give us some clearer signposting.
Cora Hoexter is a Professor in the School of Law at the University of the Witwatersrand, Johannesburg, and is currently President of the Administrative Justice Association of South Africa. She has written extensively on administrative law and is the author of a leading scholarly text, Administrative Law in South Africa (Juta & Co Ltd, 2 ed 2012).
 For a detailed recent account of this contest and its implications, see Melanie Murcott & Werner van der Westhuizen ‘The Ebb and Flow of the Application of the Principle of Subsidiarity: Critical Reflections on Motau and My Vote Counts’ (2015) VII Constitutional Court Review (forthcoming at www.ccr.org.za).
 Section 33 provides:
‘(1) Everyone has the right to administrative action that is lawful, reasonable and procedurally fair.
(2) Everyone whose rights have been adversely affected by administrative action has the right to be given written reasons.
(3) National legislation must be enacted to give effect to these rights, and must–
(a) provide for the review of administrative action by a court or, where appropriate, an independent and impartial tribunal;
(b) impose a duty on the state to give effect to the rights in subsections (1) and (2); and
(c) promote an efficient administration.’
 The SCA is an intermediate court of appeal between the High Court and the Constitutional Court.
 The problem of officials seeking review for dishonourable motives is becoming acute in South Africa: see Leo Boonzaier ‘Good Reviews, Bad Actors: The Constitutional Court’s Procedural Drama’ (2015) VII Constitutional Court Review (forthcoming at www.ccr.org.za).