Constitutional implications of Covid-19:
The striking down of the lockdown regulations
Jason Brickhill
This new contribution seeks to provide a weekly analysis of constitutional issues arising from Covid-19 and the responses to it. In this instalment, I consider the judgment in De Beer striking down the lockdown regulations.
On 2 June 2020, the North Gauteng High Court struck down the ‘lockdown regulations’ in De Beer and Others v Minister of Co-operative Governance and Traditional Affairs [2020] ZAGPPHC 184. The judgment is deeply flawed. It appears to be the result of an application lacking in rigour and defended without adequate rigour. Urgent court often lowers the standards of care with which cases are litigated and judgments prepared, but this ought not to happen in a matter of such importance.
The papers in the application are not yet publicly available, so it is only possible to understand the attack on the regulations, and the defence mounted in response, based on the judgment itself. However, this much is clear (para 3.1): the applicants challenged the declaration of a state of disaster and the regulations as a whole; they sought to allow all gatherings, subject to conditions. In the alternative, the applicants sought an order allowing all businesses to operate subject to safety conditions. In response, the Minister of Co-operative Governance and Traditional Affairs appears to have mounted a defence to the scheme of the regulations as a whole.
The central flaws in the judgment are: (1) the court fails to test each regulation individually, (2) it applies a proportionality approach rather than rationality, and (3) it muddles rationality review and limitations analysis.
The applicants, the scope of the challenge and procedure
Before I address each of these problems with the judgment, a brief comment on the applicants is appropriate. The application was brought by Reyno De Beer, an attorney, and a voluntary association known as Liberty Fighters Network (LFN). LFN appears to be an alter ego of De Beer, whose contact details are listed on its website. A non-profit organisation called ‘Hola Bon Renaissance Foundation’ was admitted as amicus curiae.
Not much is known about LFN. From its website, it appears to be campaigning on two issues – challenging the lockdown and ‘stopping evictions’. It also appears to be an organisation committed to libertarian values and, based on ‘news’ content posted on the website, to have an interest in Zionism and Donald Trump.
LFN was represented in the hearing by Zehir Omar, an attorney. Omar is no stranger to controversial litigation. He acted for the applicant in Omar v Government of the Republic of South Africa [2005] ZACC 17, a constitutional challenge to s 8 of the Domestic Violence Act 116 of 1998. The challenge was directed at the mandatory requirement to issue a warrant of arrest when a protection order is issued for domestic violence. The Constitutional Court dismissed that application.
The amicus curiae, the Hola Bon Renaissance Foundation (HBR), was the entity that initially attempted to challenge the lockdown regulations in an application for direct access to the Constitutional Court, which was dismissed. Little is publicly available about HBR, which describes itself as being committed to socioeconomic transformation. In 2013, HBR issued a public apology, purportedly on behalf of all civil society organisations in South Africa, to the Gupta family following the controversy relating to the Guptas’ use of the Waterkloof Military Airport to bring wedding guests to South Africa.
There is undoubtedly an important political debate to be had about all these actors invoking the public interest, and about a white attorney and a libertarian organisation (LFN) purporting to invoke the interests of the majority of poor, black South Africans to challenge the lockdown regulations. However, I limit myself to the legal implications of who brought the challenge and how.
To begin with, the applicants undoubtedly did have standing to challenge the lockdown regulations under s 38 of the Constitution of the Republic of South Africa, 1996 – certainly in their own interest, at least, and possibly in the public interest. In addition, in light of the doctrine of objective constitutional invalidity, the applicants were entitled to argue that the regulations are unconstitutional not simply because of the effects on the applicants themselves, but because of the effect on other people’s rights (Ferreira v Levin NO [1995] ZACC 13). Indeed, the court is required to consider the full, objective position. However, it is also the duty of the court to ensure that the interests of persons not before the court are adequately protected and investigated. Where an applicant acts in the public interest and purports to rely on effects of a law on people who are not before the court, the court needs to exercise care to ensure that all the facts are fairly and adequately placed before the court.
A crucial procedural rule exists to ensure that important constitutional litigation is brought to the attention of the public, so that interested parties are able to intervene as parties or amici. The judgment does not make it clear whether Rule 16A was complied with in this case. There is no Rule 16A notice for this case on the SAFLII website where such notices may be posted. The Rule requires a notice to be published on a notice board at court identifying the constitutional issues raised in a matter. One amicus (HBR) did intervene, but it is unclear if this was prompted by a Rule 16A notice. Fortunately, the judgment has been widely publicised and any parties with an interest in the appeal proceedings will have the opportunity to seek admission.
I turn now to the three main problems with the court’s approach to the rationality review and limitations analysis – the ‘as a whole’ nature of the constitutionality inquiry; getting the rationality test wrong, and muddling rationality review and limitations analysis.
A ‘holus bolus’, shotgun approach to the challenge
The first problem with the judgment is a simple one. The applicants challenged all the level 3 and 4 lockdown regulations, and the court considered the challenge as a whole. It ought to have considered the rationality of each regulation – that is, whether each regulation is rationally related to the objective for it.
The objective of the regulations as a whole is to prevent the spread of Covid-19 so as to ‘flatten the curve’ and save lives, as the Director-General’s (DG) affidavit explained. However, individual regulations may have more specific purposes, and there may be other objectives underpinning individual regulations. A careful, individualised approach is constitutionally required, as the Constitutional Court has often reiterated.
In Shaik v Minister of Justice and Constitutional Development [2003] ZACC 24 para 25, the court emphasised that ‘[i]t constitutes sound discipline in constitutional litigation to require accuracy in the identification of statutory provisions that are attacked on the ground of their constitutional invalidity’, explaining that such accuracy is ‘especially important in those cases where a party may wish to justify a limitation of a Chapter 2 right and adduce evidence in support thereof’ (para 24).
Towards the end of the De Beer judgment, having found that ‘most’ of the regulations are invalid, the court saves a few. However, it appears to do so because it realises the risk of invalidating the lockdown regulations entirely in the context of the Covid-19 risk, not as a result of a careful legal analysis of each regulation in relation to its objective.
Getting rationality wrong
The second major flaw in the judgment is that Davis J, having initially stated the rationality test correctly, got it wrong when applying the test. Richard Calland has commented that Davis J’s approach looks more like reasonableness review than rationality. In my view, parts of the approach look like proportionality testing, the strictest review standard found in our public law.
Pierre de Vos has argued that the DG also appears to have got the law on rationality wrong. Again, we do not have the full affidavit, but the DG summed up the test as follows in the answering affidavit:
‘I am advised that in determining whether the decision of the functionary is rational, the test is objective and is whether the means justify the ends. Thus, I submit, with respect, that under the circumstances, the means justify the ends.’
As the court points out, and De Vos has discussed, the correct expression is normally ‘the end justifies the means’, and not the other way around. De Vos has pointed out that the use of the phrase ‘end justifies the means’ conveys that a good aim justifies bad methods. It is not clear from the judgment as a whole that the DG understood rationality in this way. Despite the inelegant summing up of the test, the DG seems to be identifying (correctly) that rationality is about the relationship between means and ends, that the test is objective and that the idea of ‘justification’ underpins the test.
‘Justifiability’ was introduced into rationality review by the Labour Appeal Court in the famous Carephone decision, where the court was interpreting the review standard applicable to CCMA awards. The LAC held that a CCMA award must be rationally connected to the material before the arbitrator, and that it must therefore be ‘justifiable’ in terms of the reasons given for the award (Carephone (Pty) Ltd v Marcus NO (1998) 19 ILJ 1425 (LAC) para 20). Subsequent Labour Court decisions attempted to rein in this approach to rationality as justifiability, such as Shoprite Checkers (Pty) Ltd v Ramdaw NO [2000] ZALC 27, because it blurred the distinction between review and appeal.
‘Justifiability’ may also be linked in this context to the famous concept of ‘a culture of justification’ which Etienne Mureinik argued underlies the Constitution as a whole, an idea that has influenced the development of public law review (Etienne Mureinik ‘A bridge to where? Introducing the Interim Bill of Rights’ (1994) 10 SAJHR 31–48).
All of this is to say that the notion of ‘justifiability’ has been floating around rationality review in different contexts for some time. In any event, the rationality standard applicable to challenges to regulations is emphatically not the Carephone standard of justifiability in light of the material available to the decision-maker. In summing up the test as ‘the means justify the ends’, which the DG’s counsel insisted the DG meant to say (De Beer para 6.7), the DG may well have intended to convey that the means serve the ends, and are therefore justifiable. A reading of the DG’s affidavit as a whole would confirm this. This would be, broadly speaking, a correct formulation of the test for rationality of regulations.
In any event, the court itself correctly set out the test initially (paras 6.1–6.4) but then applied an entirely different, and wrong, approach to testing the regulations.
First, Davis J adopted an approach that involved comparing one regulation to another, pointing out that some are stricter than others (for example, not allowing family members at the bedside of hospitalised patients but allowing 50 people at a funeral). This is not testing for rationality, but for proportionality or reasonableness. The court also appeared here to be plucking ‘examples’ from throughout the regulations, and weighing them against each other to test the regulations as a whole for rationality. The court’s approach seems to be that if one regulation is stricter than another that deals with similar subject-matter, both regulations are irrational. However, each of the regulations imposes restrictions on movement and gathering that, on the face of them, clearly relate to the objective of slowing the transmission of Covid-19. The fact that a regulation is stricter than another may raise other questions but does not on its own lead to irrationality.
Muddling rationality review and limitations analysis
The final problem with the court’s approach is that having set out the rationality test correctly in the body of the judgment, when it comes to applying the test to the impugned regulations, it begins to apply a hybrid rationality-limitations analysis that leads it badly astray. This is most glaringly apparent from paragraphs 9.3 and 9.4 of the judgment when the court sets out its conclusions. It is worth quoting these paragraphs in full:
‘9.3 In every instance where “means” are implemented by executive authority in order to obtain a specific outcome an evaluative exercise must be taken insofar as those “means” may encroach on a Constitutional right, to determine whether such encroachment is justifiable. Without conducting such an enquiry, the enforcement of such means, even in a bona fide attempt to attain a legitimate end, would be arbitrary and unlawful.
9.4 Insofar as the “lockdown regulations” do not satisfy the “rationality test”, their encroachment on and limitation of rights guaranteed in the Bill of Rights contained in the Constitution are not justifiable in an open and democratic society based on human dignity, equality, and freedom as contemplated in Section 36 of the Constitution.’
In paragraph 9.3, the court seems to blur the ‘means-ends’ rationality question into the question of whether the limitation of any right is justifiable under s 36. In paragraph 9.4, the court explicitly holds that if the regulations (as a whole) do not satisfy rationality, then they also fail under s 36. This approach is flawed. The question of whether a regulation is irrational and whether it unjustifiably limits one or more rights in the Bill of Rights are distinct and involve entirely different approaches. Vuyani Ngalwana, who otherwise is sympathetic to the decision in De Beer because of its emphasis on individual liberty over state ‘paternalism’, observes that the court failed to distinguish the rationality and limitations inquiries.
Section 36 only permits this if it is affected by a law of general application and is reasonable and justifiable. The requirement of a law of general application serves the rule of law. It ensures that government officials may not impose new restrictions on rights simply by adopting a policy or instituting a new practice not authorised in law. Here, limitations are in the form of regulations. As Davis J points out in De Beer, regulations are made without the involvement or oversight of Parliament that would take place for legislation. However, regulations (and other forms of delegated legislation) constitute laws of general application and may limit rights, subject to s 36 (Larbi-Odam v MEC for Education (North-West Province) [1997] ZACC 16 para 27).
As I argued at the start of lockdown, the limitations inquiry is not a one-size-fits-all approach where the ubiquitous threat of Covid-19 automatically justifies the web of new regulations. Ultimately, every strand in the web must satisfy s 36. However, it also importantly follows that it is not permissible for an applicant to challenge the regulations as a whole, invoking a range of constitutional rights in this generalised, shotgun approach adopted by the applicants in De Beer.
The various regulations limit specific constitutional rights in very different ways. It is for the applicants to plead the constitutional challenge with sufficient precision to enable the government to respond accordingly, as explained above. The applicants should also adduce evidence in support of the challenge. Ultimately, of course, the onus lies on the state to justify a limitation under s 36 and, if the justification depends on facts, to put up the necessary evidence (Minister of Home Affairs v NICRO [2004] ZACC 10 para 36).
However, the framework for the limitations analysis is set by how the challenge is pleaded by the applicants – which specific rights are relied upon to challenge which regulation, and, ultimately, is the limitation of those rights proportionate in terms of s 36? The court in De Beer does not conduct this inquiry, instead subsuming the limitations analysis in the (flawed) rationality review that it has conducted.
Conclusion
De Beer got the rationality review of the regulations completely wrong and barely conducted a rights limitations analysis. It will, in all likelihood, be overturned on appeal. It is also likely that other actors will intervene in the appeal proceedings, which will be conducted away from the challenges of urgent High Court proceedings.
Importantly, pending appeal proceedings, it is worth emphasising that the judgment is suspended and the lockdown regulations remain fully in place. The striking down of regulations does not require confirmation by the Constitutional Court (Mdodana v Premier of the Eastern Cape [2014] ZACC 7 para 23). However, Davis J suspended the order of invalidity for 14 days and, even after that time, the effect of the order will be suspended by an application for leave to appeal in an ordinary way, barring the unlikely grant of interim execution order. The bottom line is that the lockdown level 3 and 4 regulations remain in force and binding at present.
Jason Brickhill is an advocate, author and activist. He teaches human rights law at the University of Oxford, where he is completing a DPhil in law as a Clarendon Scholar and is Research Director of the Oxford Human Rights Hub. He is an advocate at the Johannesburg Bar and the former Director of the Constitutional Litigation Unit of the Legal Resources Centre. His publications include Public Interest Litigation in South Africa (2018) and the co-authored work Constitutional Litigation (2013).
Click here to download the PDF version.
Disclaimer:
This article is provided for informational purposes only and not for the purpose of providing legal advice. The views expressed herein do not necessarily reflect the views of Juta & Company (Pty) Ltd. Juta and Company (Pty) Ltd are not liable for any damage arising should the information be followed without proper due process being undertaken by the reader. Before making any decision or taking any action based on the information contained herein, which decision or action might affect your personal finances or business, you should consult a qualified professional advisor.
