(1) While the intense concentration of people in a given location (such as Luanda) can produce problems, the people themselves are not and are never the problem. I think this is an extremely important distinction: one phrasing (”people are the problem”) implies that getting rid of people is the solution (an act of incredible rhetorical violence which is capable of producing actual violence), while the other (”the concentration of people can produce problems”) allows us to admit that we need solutions without being cavalier about the humanity of the people whom we speak of. I don’t want to assume that you mean the former, but I want to make this point regardless of what you meant.

(2) That said, I’m not sure that “should Luanda grow” is a very important question for architects to address, as it would be purely rhetorical. What I mean is that, even though our study of fog farming is entirely hypothetical, it is the sort of hypothetical project which could be realized through architectural practice: these nets could be refined in design, could be built, could be funded through the mixture of means we have suggested, and could be placed within the city by the mixture of actors we have suggested. They interact with a problem — the lack of clean water — which is real and suggest a means by which architectural practice could address the problem. While there may or may not be more people concentrated in Luanda than the ecology of the region can support, it is hard for me to see how overconcentration is a problem which landscape/architecture could address.

(3) Finally, I think Stephen is right that it is very difficult to assert that Luanda is too large when its infrastructure is so clearly lacking in ways which we have the technological tools to address. When I see a photo of a street in the musseques serving as both sewer and street, I don’t think: “too many people!” — I think: “this is a problem which improvements in infrastructure could address”. I’ll certainly allow that its possible that, even with the deployment of every technological tool at our disposal, Luanda still would not be able to provide adequately for the needs of its population — but I don’t think we have anything like the kind of information necessary to say that definitively. Whereas we do have the information to say “if we made these improvements to the infrastructure of the city, we would improve the lives of the inhabitants in these ways”. Which is enough for me.

The amount of population a city can maintain has always been linked to technology and infrastructural sophistication. It was farming technology which caused the development of cities in the first place. Sewage technology resolved for London many of the sanitation issues now felt by Luanda, allowing that city grow while sustaining a previously unsustainable population. I was in NYC yesterday, and it would have been a horrible experience if that city only had Luandan technology.

Your concern that the growth of Luanda will always outpace it’s water infrastructure (among others) is vital however, and I don’t have an answer for it. But I do believe an answer exists, and it may require the sort of temporary population tempering you advocate to allow population and technology to grow in synch; with the minimum human suffering possible.

I am not advocating a reliance on the ‘natural’ constraints to unsustainable growth - war and disease - but I fear that water riots and cholera will continue to be a feature of a city where extra water infrastructure never quite keeps up with population growth.

In short, the problem isn’t water, it’s people: and water shortages are a symptom which will not be cured without seeking humane ways of reducing the population in Luanda to sustainable levels.