(constitutionhill.org.za)
Dan Mafora’s repudiation of the plain meaning of the Constitution and the accepted rules of legal interpretation in his “Expropriation Act: Van Staden misrepresents the Constitution and the law” generates worry.
In fact, there is so much to say, that I have been forced to limit my response to only three of his most egregious submissions.
Legal fictions
Mafora is entirely correct that legal fictions are acceptable jurisprudential institutions. But he stretches their scope too far.
Sections 25(2) and (3) of the Constitution require the “payment” of an “amount” of “compensation.” A formalistic appeal to “nil compensation” does not satisfy one of these three elements, and purporting to do so is simulation.
Mafora agrees, writing, “This would be true if it was the full story,” with the proviso that he believes I am not recognising the validity of legal fictions.
The Expropriation Act comes after the Constitution and its requirement for “payment” of an “amount” of “compensation,” and sits downstream from it.
If we are to suppose that ordinary legislation – controlled and governed by constitutional protections and rules – may introduce legal fictions regarding those controls, then the whole edifice of constitutional supremacy crumbles.
Allow me to sketch another example:
Section 49(1) of the Constitution provides that Parliament is elected for a term of “five years.” Suppose that Parliament adopts a Term Limits Act which, among other things, defines “one year” as “ten sets of 365 days” – in other words, one year is 10 years, and five years is 50 years. Thus, by constructing a legal fiction in ordinary legislation concerning a constitutional rule, Parliament would unilaterally extend its own term by decades.
This is not difficult stuff.
In systems of constitutional supremacy, constitutions may introduce legal fictions – the presumption of innocence comes to mind – but legal fictions may not be introduced into constitutions, like the Expropriation Act does. To suppose otherwise would necessarily invalidate constitutional supremacy per se.
If Mafora’s argument is to be accepted, there is no reason to ever waste time on “amending” the Constitution ever again, if Parliament can simply fictionalise provisions via the route of ordinary legislation. In fact, we may as well chuck the Constitution entirely.
Grotius
Mafora goes on to explain that Hugo Grotius’ original conceptualisation of expropriation does not necessarily require compensation, because Grotius added a proviso of “[compensation] if possible”.
We are in the advantageous position that even if so, this is inapplicable in South Africa primarily for two reasons.
The first is that there is no question of the state being unable to afford to compensate owners of property for expropriation.
Grotius notes “if possible,” not “if desirable.”
For the 2025/26 budget, government allocated some R10 billion to land reform, still less than the R12 billion allocated to sports, arts, and culture, and the collapsed Road Accident Fund’s R30 billion. I guess this is what the “land crisis” is worth?
According to the South African Institute of Chartered Accountants, South Africa lost R1.5 trillion to corruption between 2014 and 2019 alone.
At the time of writing, a 900-hectare farm in Fouriesburg in the Free State was being sold for R13 million. With just the money lost to corruption, the South African government could have purchased 115,384 of these farms on the open market.
According to StatsSA, in 2017 there were only 40,122 farming units in the country totalling some 46.4 million hectares.
In fact, even if one were to price all such land according to the highest provincial average price per hectare, namely Gauteng (at almost R11,000 per hectare), then one would still require roughly a third of the money lost during the so-called “state capture” period to completely transform landownership patterns in one fell swoop, without breaking down property rights.
The second reason is that even if the nil compensation fiction is to be accepted – and it cannot be – the pretence of “paying” this fictious “amount” of “compensation” amounts unambiguously to a limitation of the right to compensation, and thus activates section 36 of the Constitution – the limitations provision.
Part of the analysis when this protective clause of our Constitution is brought to bear is to consider whether there are “less restrictive means” available to the state to achieve its legitimate ends, other than the rights-limiting conduct in question.
According to the “if-possible” criterion, the less restrictive means is simply to wait until it becomes possible (SARS is always hard at work!) to pay market compensation. The only exception imaginable here would be a state of emergency, in terms of which section 25 of the Constitution would in any event be suspended.
Smoke and mirrors
Section 25(8) of the Constitution provides that, “No provision of this section may impede the state from taking legislative and other measures to achieve land, water and related reform, in order to redress the results of past racial discrimination, provided that any departure from the provisions of this section is in accordance with the provisions of section 36(1).”
Perhaps the most concerning argument Mafora makes is, on the strength of this provision, to submit that “Parliament has the power to deviate from provisions of section 25”.
If Mafora’s argument about legal fictions implicitly nullify constitutional supremacy, he does so explicitly here, making the legally unsustainable submission that section 25(8) renders the remainder of section 25 effectively nugatory.
I assure readers that neither Mafora nor I, who attended exactly the same constitutional and property law classes, at the same time, at the same law faculty, were taught such tinpot counter-constitutional nonsense.
In fact, we were taught the precise converse.
Two of the basic rules of legal interpretation are, firstly, that a statute may not be interpreted as containing invalid or purposeless provisions; and secondly, where there exists more than one potential meaning of a legal provision, the courts do and must prefer the construction that is less harsh, inequitable, or unreasonable to the legal subject.
Where these rules apply to contracts and legislation, they apply doubly to constitutional instruments. A supreme constitution can obviously not be read as containing throat-clearing words without effect, nor can its words – particularly the words in the bill of rights – be given the most conceivably unreasonable, inequitable, and rights-defeating interpretations.
What, then, is section 25(8) about?
During the constitutional transition, many left-wing activists and jurists, who had significant intellectual sway on the negotiations, worried that under a constitution with strong property protections a future government would shy away from redressing the injustices of the past.
Within this context, and understanding the rules of legal interpretation, and the impermissibility of rendering whole sections of the Constitution simply inoperative, it is clear that section 25(8) is a clarificatory provision that makes clear that government must engage in land reform and may not, through inaction, simply hide behind section 25(1).
Moreover – and crucially – section 25(8) involves section 36 in the determination of the justifiability of property-rights-limiting actions of government in pursuance of land reform.
This is something, for example, that the Constitutional Court has (errantly) tried to make impossible in its 2004 Van Heerden affirmative action judgment.
Briefly, in terms of section 9 of the Constitution, and the preceding 1997 Harksen judgment, if discrimination occurred on a prohibited ground (like race), a presumption of “unfair” discrimination has been established, and it falls to the discriminator to justify the discrimination in terms of section 36.
Van Heerden turned this crucial protection on its head, with the court bizarrely concluding that the question of fairness or unfairness does not arise if the given discrimination satisfies the requirements of section 9(2) read in isolation. This not only rendered the test for fair or unfair discrimination in affirmative action cases null and void, but closed the door to a section 36 limitations analysis.
Section 25(8) makes this curbing of constitutional rights impossible in the case of section 25, necessitating a section 36 analysis. (This needs clarifying for many: section 36 is in the bill of rights, and is part of the regime of rights-protection, not rights-violation as many authoritarians would have it.)
Jurisprudential coherence
Mafora’s insulting send-off is unfortunate, but unsurprising after such an uninspiring and unsuccessful attempt to rally a coherent legal argument.
Ultimately, readers need to decide whether the formal supremacy of the Constitution is something meaningful or useful – which I submit it is – or whether, on Mafora’s account, government can either use clever wording to sidestep constitutional protections, or simply “deviate” at a whim.
Martin van Staden is the Head of Policy at the Free Market Foundation. He has an LL.M. (cum laude) from the University of Pretoria, where he is presently pursuing an LL.D. His work on expropriation without compensation has appeared in, among others, the Pretoria Student Law Review, the Potchefstroom Electronic Law Journal, and the International Property Rights Index.