The decision by South Africa's Constitutional Court in Mazibuko and Others v. City of Johannesburg and Others, handed down in October 2009, highlights the perplexing issues entailed in recognizing a constitutional right to water. Here's a preliminary question that's not so perplexing: if people have a right to water, is it a right to unlimited water? The answer must be no, since there's no such thing as unlimited water, least of all in a "largely arid country" (as the Constitutional Court characterizes South Africa, paragraph 3). But here's an observation that makes this first proposition less helpful than it might have seemed: for someone who has plenty of water to make judgments about how much water other far less fortunate people are constitutionally entitled to is deeply discomfiting. One must always empathize with people who are so profoundly in need -- and yet South Africa cannot provide all that its people need.
But if no one has a right to unlimited water, then can water be automatically cut off after a certain amount has been provided? This was one of the issues in Mazibuko. It's worth emphasizing that this question in a sense only could arise because something quite impressive already existed in the city of Johannesburg, namely a system that did provide some amount of water to each person, and without charge, as a matter of constitutional right.
So the question really was, what happens when this guaranteed, free water runs out? Even if we agree that the water can't just keep flowing forever, no matter what, a seemingly technical legal question may have great practical significance: if the water shuts off, does the shut-off count as a deprivation of a right or not?
If you have a right to water, subject to cut-off when that step is properly justified, then arguably you have a right to a hearing each time you run over your allotment -- say, around day 25 of each month, if your allotment is distributed on a monthly basis. If, on the other hand, your only right is to the basic allotment, then when the meter runs out the water simply stops, without any further ado.
The word "meter" isn't metaphorical. The system the Constitutional Court considered, and unanimously upheld, included pre-paid meters which did indeed turn off the water (unless additional water credits were purchased) when the monthly allocation ran out.
Constitutional abstractions aside, the problem is a stark one: protecting each person's access to more water if they rightfully should receive it is likely to be costly (in terms of hearings held, and water supply continued) and likely also to encourage some people who rightfully shouldn't get extra water to game the system -- all to the detriment of the overall societal effort to supply water and other socioeconomic rights to all. But cutting off water will harm some truly needy people -- even, as the South African legal scholar Michael Kidd has noted, leading to outbreak of disease as cut-off users turn to alternate, unsafe sources of water.
It seems to me that the Constitutional Court essentially concluded that this dilemma had no perfect solution, and that it would therefore defer to the Johannesburg water authorities, who were at least making sincere and diligent efforts to fulfill their task of providing water to all. There is good reason for such deference. Courts can't run water systems, and if the design and operation of water systems inevitably involves choices among various imperfect alternatives, those are the sorts of choices we generally think elected decisionmakers should make.
But it is very troubling to think of families turning in desperation to tainted water. The Johannesburg system allocated water by the "stand" (a unit of property), and actual stands have widely varying numbers of people living on them. No doubt the poorest stands tend to be the most crowded, and for the people on these stands the water allotment per person is at its least. Specifically, Johannesburg allocated 6 kilolitres of free water to each stand, an amount that supplied each resident of a stand with an average household size (3.2 people) with approximately 60 litres of water per day. That amount, the Constitutional Court noted, was well above what even the challengers of the policy said was constitutionally required. (Paragraph 88) But for a household in Phiri (an area of Soweto), where an average of 8.8 people lived on each stand, this policy would deliver only about 23 litres per day per person, a worryingly low amount.
These poor households were not entirely without recourse, since a program allocated 4 extra kilolitres per household if they registered as indigent. I agree with the Constitutional Court that requiring registration as indigent is not unacceptably stigmatizing (paragraph 101), but in fact the program apparently didn't work very well, since only a little over 1/5 of the eligible households registered. (Paragraph 81)
Perhaps some better distribution system should have been required, to insure, or at least make it more likely, that no one faced acute health risks as a result of a water cut-off. Increased allocations for specially crowded stands might have been feasible, though the Court thought otherwise (paragraph 84). Or some system of emergency aid might have been put in place. But a perfect guarantee seems very hard to achieve, unless the water never stops flowing -- and so all policies that do cut off water are balancing individual against community need.
It's also conceivable, however, that in Johannesburg there were enough safe alternative sources of water so that the cut-off system did not pose the same acute health risks as it could have elsewhere. And it must be said that while the poorest of the poor were at risk of shut-off, they were not being deprived of, or denied, all water -- they received the basic, free allocation. In that respect, this case is weaker than Government of the Republic of South Africa and Others v. Grootboom and Others, the pathbreaking 2000 decision in which the Constitutional Court decided that a plan to implement the constitutional right to housing had to make some provision for the needs of the neediest, those homeless here and now.
To all this, one further complication must be added. In the last years of apartheid, a culture of nonpayment had grown up in Soweto and other predominantly black areas of Johannesburg and South Africa. Nonpayment was a form of resistance to apartheid, but it did not end with the achievement of democratic government. So a further problem for the Johannesburg water authorities was that people were accustomed to getting water without paying for it.
How to overcome a culture of nonpayment? That's another difficult question. One wrong answer would be by brute force and confrontation. The people are not to be bludgeoned. Yet who would not prefer free water to water available only for a fee? Men and women are not angels, and so high principles alone will not likely alter a practice so seemingly in one's self-interest (only seemingly, since in the long run water costs society money, and will not arrive without it).
So the pre-paid meter system becomes very attractive. It switches the burden of inertia: now, without payment, the water will automatically run out. And it individualizes the situation: just your water runs out, since you didn't buy water credits. In a society with a culture of payment such a system would be unnecessary, but perhaps in South Africa it was called for. That wouldn't remove the need for measures to protect those who ran out of water and could not afford to pay for it, but it might justify using the pre-paid meter system in the first place.
Finally, what should one make of the evidence cited by the Court to the effect that after people got used to the new system, they were generally rather happy with it? (Paragraph 18) The installation of the meters went in tandem with upgrading the Soweto water supply infrastructure, and it's plausible to think the two really were dependent on each other -- no dependable revenues, no way to finance a new infrastructure. And perhaps that is what most of the people of Phiri concluded.
If that's so, does it matter constitutionally? In some contexts, we would say no. Even if many people want my right to free speech abridged, I'm still entitled to exercise it. (Though not always and everywhere -- not with a loudspeaker in a quiet neighborhood late at night, for instance.) But socioeconomic rights seem more intrinsically concerned with costs: if I am entitled only to those rights that the state is progressively achieving by reasonable measures within the limits of its available resources (the characteristic dimensions of the socioeconomic rights in South Africa's constitution), then reasonable choices about who gets what seem intrinsic to what the constitution secures. Unless there is some minimum amount that everyone must have -- and that the state can in fact provide -- there will inevitably have to be balancing of the socioeconomic claims of some against others.
And if that balance is to be made, it seems to me that the judgments of the people themselves about whether that balance has been struck rightly are relevant. The residents of Phiri who were satisfied did not have the right to deprive their unhappy neighbors of constitutional protection -- but their views were, I think, relevant to deciding whether that protection, the guarantee of reasonable measures to provide water, had been provided.
So, with the privilege of as much water as I want, I hesitantly suggest that the system of pre-paid meters in Phiri, Johannesburg was constitutional in and of itself, though the constitution might have required more than Johannesburg had done to ensure that the neediest of its families had enough water to meet their basic human needs. I welcome comments by those closer to the facts about whether they agree.