Dave Pollard asserts, as one of his three "premises that have upset my readers the most" in his environmental philosophy that:

That humans are not meant to live in cities or other crowded habitats, and that any utopian society must allow and encourage people to spread out and live in close contact with the rest of nature;

I can agree with a weaker version of this point on an ecological level. I see it as a counter to the idea that we should sacrifice a few areas to be high-density cities and let everything in between revert to wilderness -- an outgrowth of the idea that humans inherently pollute nature that I criticized before. But interestingly, much of the commentary on this point in his comment thread dealt with the social aspect of city versus small-town life (a common enough discussion that I won't limit myself to the points made by Pollard's commenters).

Proponents of small towns praise the "you know everyone you see" aspect, while deploring the anonymity of the city. City-lovers respond by talking about the excitement and vibrancy of the city versus how boring small towns can be. But what both seem to like is a density of social interactions. Both see their less-preferred scenario as socially anemic -- a city full of strangers, or a town full of the same old people doing the same old things.

The difference may perhaps be conceived of in terms of organic versus mechanical-supplemented-by-familial solidarity. Fans of the city tend to thrive on a wealth of social functions -- clubs, museums, shops, concerts, etc. It's the scenarios and the roles, more than the people in them, that get them hooked. Fans of small towns, on the other hand, see more value in establishing social connections (though not necessarily kinship ones, particularly in the case of left-leaning people) with particular individuals.

To truly argue for preferring (on a policy, rather than personal, level) one size settlement, then, requires attention to the type, rather than just the ammount*, of social interaction.

*I originally wrote "quality, rather than just the quantity" here, but that seemed misleading. We often think of quality as something that can be measured on a scale -- i.e., as a quantitative measurement. This is a useful notion when talking about the value within an item versus the number of items, especially when it allows a clear tradeoff between the two (e.g. 20 sheets of 2-ply toilet paper versus 40 sheets of 1-ply). But it's confusing when you're trying to indicate the idea of a non-ranked (or not-yet-ranked) difference of kind or type.