I imagine it's partly that 'district' combines the sound 'dys' with the word 'strict,' so we get a bad kind of strictness suggested along (perhaps) with the German sound of rict, as in Recht and Richter. All very fascistic, in other words. But it's also just a vague term, like 'zone', and hence unhomely, anonymous. Adding a number to it makes it even less personal. And this made me think about names and home and the Spielraum that Kraus says is provided for culture by the distinction between the urn and the chamber pot. Dividing the places where people live into numbered districts with no other names provides no such Spielraum. You can love (or hate) England or Virginia, but it would surely be hard to love a numbered district. Or perhaps rather: for us a name like "District 12" sounds much too bureaucratic, impersonal, ahistorical, unpoetic to be the name of anything lovable. (There's no reason why such a name could not acquire associations and a history, and become, say, "the Fighting 12th.") It's an inhuman kind of name, in short. Which brings me to Wittgenstein and G. A. Cohen's conservatism.
The Investigations begins with a quotation from Augustine that is oddly, or strikingly, generic or neutral: "When they (my elders) named some object, and accordingly moved towards something..." The Latin is even more generic, referring not to Augustine's elders but to 'majores hominem,' which I would translate as "big people" (Google translate says "larger men") and the German is "the adults." There is something dreamlike about these nonspecific people nonspecifically moving (neither walking nor running, for instance) towards nonspecific objects. But there is also something technical about it, like an approximation of the language of physics (a body b approaches an object o at velocity v...). This abstraction from particulars is surely part of the reason why Wittgenstein talks about Augustine giving us a picture. It is a kind of model of how we learn language, whether or not it is an accurate one. We need this kind of abstraction sometimes, but we also resist it (by not liking physics or by not wanting to live in places called things like District 12).
It seems to me that Augustine offers an approximation of the truth. I wouldn't say he was hopelessly confused or wrong, but he doesn't give us the truth, the whole truth, and nothing but the truth. He gives an interpretation of part of the truth. And when Wittgenstein says of his shopping story that, "It is in this and similar ways that one operates with words," he is doing the same kind of thing. As Stephen Mulhall points out, we don't actually operate in the way Wittgenstein describes. That is, we don't write "five red apples" on a piece of paper and hand this to someone who hands it to a shopkeeper who keeps apples of all colours in a draw then looks up "red" and counts to five to identify what he is to hand over. That never happens. But it isn't false to say that we "operate in this and similar ways." We certainly do things that are similar to this. And what is a way? To behave in the same way is to do something similar, the same but different. And how we shop is similar to but different from what Wittgenstein describes. There is a certain Spielraum in the concept 'way.' We act in that general kind of way, but not in that exact way.
There is something alien about the shopkeeper, though, as Mulhall also points out (he uses the words 'robotic,' 'homunculus,' and 'foreign'). There is something equally alien, I think, about the Tractarian idea of precisely defined possibilities and definite facts. The human world is not so precise or anonymous. Red apples, typically, only need to be roughly red. That is, they can be a bit green in parts, and they don't have to be held up against a colour chart the way that a specific shade of paint might be. We don't require that kind of accuracy. Something sufficiently similar to the standard will do. The same but different. And knowing what is similar enough, what is too different to count, takes judgment.
We don't care only about objects as bearers of certain particular qualities either. We also care about particular objects. This is an idea that Cohen defends. And I don't think it's just a coincidence that "Maximizing consequentialism stands in especially sharp contrast with conservatism as [he has] defined it." That is, I think there's something of a theme in Wittgenstein that Anscombe picked up on in attacking consequentialism. Not that Wittgenstein attacks consequentialism, even implicitly, but he helps clear the way for such an attack.
So how do I get from the imprecision of (what I'm calling) the human to the rejection of consequentialism? Imprecision in definition gives wiggle room to the use and hence meaning of words. And that means we can't map words precisely to sets of definite qualities. So what we call 'good' depends on the use of judgment. We can call good only happiness or what produces happiness. But we don't have to. Even if everything we recognize as good produces happiness, we can still apply the term in (similar but) different ways in future. Ethics need not be about particular properties or their maximization. It can be about things, and the meaning of things. We can oppose using an urn as a chamber pot, for instance. Wittgenstein's shopkeeper would have a hard time doing this. He opens the draw of containers, finds one that matches the shape he has been shown, and hands over the round thing with the open top. Could he have some marked as only for this or that use? Yes, of course. But there's nothing in the picture Wittgenstein gives us to suggest that he would do such a thing. He has no culture.
Perhaps that's all hopelessly obscure, or just hopeless. But Cohen's paper is worth reading. Or listening to. The video of his talk is here: